OS2 World Community Forum

OS/2, eCS & ArcaOS - Technical => Applications => Topic started by: Mark Szkolnicki on April 09, 2018, 07:52:29 pm

Title: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 09, 2018, 07:52:29 pm
Hi All!

Thought I'd start a new topic related to testing the new VirtualBox development drop, both successes and failures.

I did some more testing on the weekend.

I have two physical optical drives in this machine (Drives S: and T:, DVD and BluRay), and was able to mount either one or the other when the Windows XP VM is active, using Drives | Optical Drives | Host Drive S: or Host Drive T:. I was able to download files, and install programs successfully using either mounted drive.

One thing I did note. After shutting down the VM, I removed the CD from one of the drives and attempted to start  up the VM again. It initially started with
the opening window, but after about 5-10 seconds later it aborted with an error window opening:

"PIIX3 cannot attach drive to Secondary Master VER_IO_NOT_READY".

I've attached a JPG of the error below.

It seems that a disk (any disk) must be in the mounted drive to restart the VM again.

I tried the ejection signal from Windows XP on the disk, while the VM was running, which was successful and then, shutting down and restarting the VM, got the same startup error.

If you are not planning to keep the disk in the drive, I found the correct sequence seems to be Drives | Optical Drives | Remove Disk From Virtual Drive, prior to closing, which clears any of the drives and ISO's that are mounted.

Don't know if that's a feature common to all versions of VirtualBox on all platforms but haven't run into that with VPC, which I'm very familiar with.

Also. as a test, created a VDI based Windows 2000 VM as I happened to have one lying around. The virtualization was extremely slow (took about 3 hours to install the OS, while I was doing other things. Again don't know if that is typical of installation on VBox or not, but worth noting.

Best to all!

Mark

Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 09, 2018, 09:56:40 pm

Also. as a test, created a VDI based Windows 2000 VM as I happened to have one lying around. The virtualization was extremely slow (took about 3 hours to install the OS, while I was doing other things. Again don't know if that is typical of installation on VBox or not, but worth noting.



I think I found the problem related to the virtualization being slow, as per above.

I created two new VM's on the weekend (Windows 98, Windows 2000) but did not check the settings.

It seems "enable VT-x / AMD-V" are enabled by default in the settings. I was not sure about that previously in my posts to Valery in the thread "VirtualBox Problems" as the original VM I had created (Windows XP) was done under Windows 7 and Windows VirtualBox v5.0.22, so I wasn't sure if I had fiddled with the settings at some point.

These two brand new unused VM's both had "enable VT-x / AMD-V" checked, when I reviewed the settings, which can cause problems in OS/2 / eCS / ArcaOS based systems (not OS/4)

I've since disabled them, as per Valery's instructions previously. I suspect that will show a marked improvement in install times, as I plan to try installing Windows 2000 again (ran into problems the first time).

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 09, 2018, 10:38:43 pm
> "PIIX3 cannot attach drive to Secondary Master VER_IO_NOT_READY".

Did you removed the CD from your CD/DVD drive? If so, insert it back, or detach from "Media" menu. Note that Physical media support is unstable, yet. For example, you can easily fail to boot from physical media.

> Also. as a test, created a VDI based Windows 2000 VM as I happened to have one lying around. The virtualization was extremely slow (took about 3 hours to install the OS, while I was doing other things. Again don't know if that is typical of installation on VBox or not, but worth noting.

You need to use HW virtualization, to run heavy-weight OS-es (but it currently does not work on IBM kernels, so OS/4 kernels only for now)

> It seems "enable VT-x / AMD-V" are enabled by default in the settings. I was not sure about that previously in my posts to Valery in the thread "VirtualBox Problems" as the original VM I had created (Windows XP) was done under Windows 7 and Windows VirtualBox v5.0.22, so I wasn't sure if I had fiddled with the settings at some point.

No, I just created the new VM from scratch, VT-x/AMD-V is disabled. In any case, disable it if it enabled, for some weird reason (unless you have one of the latest OS/4 kernels).

win2000 may run ok without VT-x enabled, but win7 or ubuntu are slow like hell, and work conveniently only with VT-x enabled

PS: Why did you created the second topic? Is not the previous one good enough?
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 09, 2018, 10:47:34 pm
Hi Valery!

Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 09, 2018, 11:17:29 pm
2Mark Szkolnicki:

Ah? Yes, just tried on an IBM kernel: It enables AMD-V/VT-x by default. Not sure why, but ok, just disable it for now.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 09, 2018, 11:55:12 pm
Hi Valery!

> "PIIX3 cannot attach drive to Secondary Master VER_IO_NOT_READY".

Did you removed the CD from your CD/DVD drive? If so, insert it back, or detach from "Media" menu. Note that Physical media support is unstable, yet. For example, you can easily fail to boot from physical media.

> Also. as a test, created a VDI based Windows 2000 VM as I happened to have one lying around. The virtualization was extremely slow (took about 3 hours to install the OS, while I was doing other things. Again don't know if that is typical of installation on VBox or not, but worth noting.

You need to use HW virtualization, to run heavy-weight OS-es (but it currently does not work on IBM kernels, so OS/4 kernels only for now)

> It seems "enable VT-x / AMD-V" are enabled by default in the settings. I was not sure about that previously in my posts to Valery in the thread "VirtualBox Problems" as the original VM I had created (Windows XP) was done under Windows 7 and Windows VirtualBox v5.0.22, so I wasn't sure if I had fiddled with the settings at some point.

No, I just created the new VM from scratch, VT-x/AMD-V is disabled. In any case, disable it if it enabled, for some weird reason (unless you have one of the latest OS/4 kernels).

win2000 may run ok without VT-x enabled, but win7 or ubuntu are slow like hell, and work conveniently only with VT-x enabled


Regarding physical media support - I have not ran into any difficulties at all at this testing phase - all files and installations proceeded without any errors, and I can only report that all physical optical media has booted correctly and flawlessly, as of now, for the optical drives tested.

Regarding the tests creating VM VDI's - running the same test creating the Windows 2000 VM, once the AMD-V was disabled, installation proceeded and completed in under 40 minutes. Previously the install needed 3 hours and 20 minutes to complete. I noted that I only have an upgrade version of Windows 7 so will have to wait until a full test version that have ordered arrives.

Regarding "enable VT-x / AMD-V" - as a tester I can only report what occurs, Valery, based on the testing that is being done. The VM's created consistently have the feature enabled. Perhaps one of the other testers can confirm this occurring as well, by:

1) Creating a new VM
2) Hilighting the new VM (prior to installing the OS) and going to Settings | System | Acceleration
3) Seeing whether the feature is enabled or disabled by default.

The VirtualBox package I downloaded was vbox-os2-i386-v5.0.51.zip dated 22/3/2018 following the link Martin created. Might there be another package floating about or that you are using? Or perhaps a global setting that enables or disables this feature, causing what I have observed?

Regarding your P.S. question - The VBox topic has generated a lot of interest, based on the number of views.

I started the new thread as one of the big problems we always have in the OS/2 / eCS / ArcaOS community is getting people to test new items or contribute to the conversation. The focus should not only be on the problems which developers and the highly technical or experienced users discuss, but also how to use and test the software properly, to allow other useful or important feedback to be obtained.. Perhaps, with a little encouragement, especially with a lot of the new users coming on-board with the release of ArcaOS, We can get a lot more testing done, which would certainly help you and others advance development of what I personally consider an important piece of software that could help my business, as VirtualPC definitely needs a more robust and flexible follow-on for our community.

"Problems" in the title always get people to shy away from testing or even asking questions, thinking their questions may be dumb, untechnical or not pertinent. We will also seriously need some documentation for VirtualBox, as the solutions we are coming up with need to be captured to allow people to consistently install and use it successfully, especially if you or others coming up with new drops in future.

Hence this testing thread and Best!

Mark





Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 10, 2018, 12:16:11 am
2Mark Szkolnicki:

Ah? Yes, just tried on an IBM kernel: It enables AMD-V/VT-x by default. Not sure why, but ok, just disable it for now.


Hi Valery!

I'm not sure how many in the community are using OS/4 consistently or not at the moment.

I myself do not use it, and I suspect the majority of users here are using a version of the IBM kernel, especially with the ArcaOS release, so you may find many testing reports varying significantly from what you may see on OS/4.

I actually would like to test it (OS/4) at some point, but I suspect if something varies significantly from what you have experienced, testing on OS/4, the difference between kernels and structures is probably the culprit. (including the slowdowns I mentioned previously with AMD-V enabled on the IBM kernel or physical media being unstable).

That why wide testing could be very important, as this VirtualBox development continues.

As I said in my previous post I think we in the community need some formalized "prototype" and "final" documentation to distribute out with the OS/2 VirtualBox even if it comes in two flavours - OS/2 / eCS / ArcaOS and another for OS/4, as YMMV.

Best!

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 10, 2018, 12:37:57 am
2Mark Szkolnicki:

> Regarding physical media support - I have not ran into any difficulties at all at this testing phase - all files and installations proceeded without any errors, and I can only report that all physical optical media has booted correctly and flawlessly, as of now, for the optical drives tested.

Yes, reading the attached media from within the VM is good, once booted. But if you try to boot from a physical floppy, or physical CD, you can encounter some problems (possibly, VM will crash).

> The VirtualBox package I downloaded was vbox-os2-i386-v5.0.51.zip dated 22/3/2018 following the link Martin created. Might there be another package floating about or that you are using? Or perhaps a global setting that enables or disables this feature, causing what I have observed?

No newer version was released. so far. I'm not aware of any global version, which disables it for all VM's. Just disable it for every VM created for now. Possibly, I'll try to disable it globally if required features are missing on your kernel, in the future.

Quote
Regarding your P.S. question - The VBox topic has generated a lot of interest, based on the number of views.

I started the new thread as one of the big problems we always have in the OS/2 / eCS / ArcaOS community is getting people to test new items or contribute to the conversation. The focus should not only be on the problems which developers and the highly technical or experienced users discuss, but also how to use and test the software properly, to allow other useful or important feedback to be obtained.. Perhaps, with a little encouragement, especially with a lot of the new users coming on-board with the release of ArcaOS, We can get a lot more testing done, which would certainly help you and others advance development of what I personally consider an important piece of software that could help my business, as VirtualPC definitely needs a more robust and flexible follow-on for our community.

"Problems" in the title always get people to shy away from testing or even asking questions, thinking their questions may be dumb, untechnical or not pertinent. We will also seriously need some documentation for VirtualBox, as the solutions we are coming up with need to be captured to allow people to consistently install and use it successfully, especially if you or others coming up with new drops in future.

Hm, I don't think the word "Problems" will get people shy away from the thread. But indeed "Testing" is better name than "Problems", I agree.

PS: BTW, in the end of older thread I added some CONFIG.SYS options to try for you. You said that you had to close Seamonkey, to run VBox successfully. I'd suspect that if you try these options, you will be able to run both at the same time (if you'll make everything correctly)

Quote
Hi Valery!

I'm not sure how many in the community are using OS/4 consistently or not at the moment.

A big number, in fact.

Quote
I myself do not use it, and I suspect the majority of users here are using a version of the IBM kernel, especially with the ArcaOS release, so you may find many testing reports varying significantly from what you may see on OS/4.

I actually would like to test it (OS/4) at some point, but I suspect if something varies significantly from what you have experienced, testing on OS/4, the difference between kernels and structures is probably the culprit. (including the slowdowns I mentioned previously with AMD-V enabled on the IBM kernel or physical media being unstable).

That why wide testing could be very important, as this VirtualBox development continues.

Indeed, not so big difference. OS/4 kernels contain some enhancements. For example, consistent hi-res timer support, based on HPET timer; Special KEE's for consistently getting curent CPU number (no -1018 error), support for CPU rendez-vous (which allows to use HW virtualization, etc., etc.)

Quote
As I said in my previous post I think we in the community need some formalized "prototype" and "final" documentation to distribute out with the OS/2 VirtualBox even if it comes in two flavours - OS/2 / eCS / ArcaOS and another for OS/4, as YMMV.

I don't think that we need two separate documentation files. Only some special notes. I think, README.OS2 file is sufficient. It did not updated since version 5.0.6. It should be enhanced more, of course, and contain OS/2-specific notes. Otherwise, documentation from Sun/Oracle applies.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 10, 2018, 03:05:46 am
Hi Valery!


2Mark Szkolnicki:

PS: BTW, in the end of older thread I added some CONFIG.SYS options to try for you. You said that you had to close Seamonkey, to run VBox successfully. I'd suspect that if you try these options, you will be able to run both at the same time (if you'll make everything correctly)

Quote

I did review your post in the "Problems" thread, and it is something I plan to test in the next few days, when I get a chance.

Best!

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 10, 2018, 09:10:36 pm
Hi Valery!


PS: BTW, in the end of older thread I added some CONFIG.SYS options to try for you. You said that you had to close Seamonkey, to run VBox successfully. I'd suspect that if you try these options, you will be able to run both at the same time (if you'll make everything correctly)


I did have some time today to test modifications to my Config.sys, to see if it would free up memory, as per your suggestions as follows

VIRTUALADDRESSLIMIT=3072 (previously 2560)
THREADS=511
PROCESSES=156

Rebooting and at start-up, I found no difference, prior to running any programs, related to available memory:

Available shared memory remained at 254 Mb
Available unused total memory remained at 90% (as displayed by SysInfo 0.9.0).

Testing as per previous, with both SeaMonkey and VirtualBox open showed the "-1018" error still occurring regularly, unless I still delayed the 15 seconds, prior to starting the VM.

I don't know where a current OS/2 version of Theseus is, to allow me to do other memory tests. I have used Theseus in tests in the past, but that was WAY past (15 years?).

Regarding ifs=d:\os2\jfs.ifs /autocheck:* /cache:128000 /lazy:8,60,5 /p1 - for the moment I will hold off on this test, as I've learned in the past not to change settings on IFS's until I fully understand what may happen. This being a working business machine, formatted almost entirely as JFS and having things to do at the moment, I would prefer to test this on one of my experimental machines first.

In regards to other tests:

I installed Windows 2000 successfully on a VM yesterday and added the VirtualBox additions from 5.0.6 there as well. For people wanting to test earlier editions of Windows, the additions both work and are a great help in using the system, as especially the video is only available in 16 colours, 800 X 600 maximum. Currently running that VM at 32 bit color resolution and 1280 X 1024.

I plan more tests over the next few days, but sporadically - running your own business does require some attention to things other than testing.

Mark

Title: Re: VirtualBox v5.0.51 Testing
Post by: ivan on April 10, 2018, 09:40:05 pm
Hi Mark,

Look for the version of Theseus here http://hobbes.nmsu.edu/h-browse.php?dir=/pub/os2/util/memory
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 10, 2018, 11:43:44 pm
2Mark Szkolnicki: Changing JFS cache to a lesser value can only cause it to perform slower. So, I doubt that if will have catastrophic consequences ;) It if set too large, it can remain insufficient memory for other needs. But if set lo a lower value, then no harm. And with VAL=3072, and JFS cache not set explicitly, it will be, most probably, still insufficient memory for kernel and drivers, so, it may be some weird system behaviour. So, I don't expect anything better with VAL/PROCESSES/THREADS set, but JFS cache not set. Error -1018 will not go away until you use IBM's kernel, and I doubt that waiting 15 seconds would be of any help. Maybe, pressing "Cancel" or "Continue" will help temporarily, but I'm not sure, as I had no such error on my machines.

Available shared memory is memory in lower shared memory arena. It does not change when you change VAL. VAL can add more high shared or private memory to userland address space if VAL is moved up to the maximum of 3072 MB, so that, VBox can allocate more high private memory for VM's.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 11, 2018, 12:14:07 am
Hi Ivan!

Hi Mark,

Look for the version of Theseus here http://hobbes.nmsu.edu/h-browse.php?dir=/pub/os2/util/memory

Got it, Sir and Thanks!

M
Title: VirtualAddressLimit
Post by: Doug Clark on April 14, 2018, 10:46:59 pm
I have been following the posting on this thread, and the VirtualBox problems - especially the parts about starting up vbox and settings in CONFIG.SYS

Yesterday I set VIRTUALADDRESSLIMIT=2048, installed and started VBOX multiple times. I tried to create a new virtual machine using a *.VHD from VirtualPC. Which was not successful. But I successfully started VirtualBox.exe multiple times, and attempted to start the new VM I created multiple times.

This morning when I booted up I received the error

Unable to allocate memory for mode info cache!
Fatal error in driver. Use Alt-Ctrl-Del to reboot.

I finally resoved this by changing my VAL back to its original setting, set VIRTUALADDRESSLIMIT=1536

I could then boot up the machine, start VirtualBox.exe, and create a new virtual machine using a Windows XP CD.

I also experience an error after VirtualBox.exe is running, when I start the VM. If I wait the 15 seconds suggested by Mark (I believe) the VM will start.

I am running ARCAOS 5.02. SNAP with dual head. I ran the YUM install lines from the readme.os2 from version 5.0.6 and I do have the
SET LIBCX_HIGHMEM=2
in my CONFIG.SYS from reading the other posts, although I do not think I have that experimental LIBC package installed.

I also noticed, which I think Valery has now seen, that the default for new VMs has  VT-x / AMD-V enabled.

I did not see a way to turn off AMD-V in my BIOS.

Title: Networking in Vbox
Post by: Doug Clark on April 14, 2018, 10:55:11 pm
Does networking work in Vbox?

I created a new VM and installed Windows XP from a CD. Windows XP of course now wants me to activate it.

I have tried NAT and BRIDGE options, and cannot ping any other machine on my local network or on the web.

And the Windows activation is failing - probably because of networking.

It has been awhile since I setup networking in Windows XP so this is probably operator error on my part.

BTW: VBOX looks pretty good so far - with my very limited usage at this point.

Doug
Title: OS4 kernel?
Post by: Doug Clark on April 14, 2018, 11:00:36 pm
Valery has talked multiple times about the OS4 kernel, and I have seen it mentioned in some other places also.

What do you guys all think about the OS4 kernel? Is it available to install? It is installable - by mere mortals such as myself?

What are the downsides of switching to OS4?

Doug
Title: Re: OS4 kernel?
Post by: Dariusz Piatkowski on April 15, 2018, 12:20:46 am
Hey Doug,

Valery has talked multiple times about the OS4 kernel, and I have seen it mentioned in some other places also.

What do you guys all think about the OS4 kernel? Is it available to install? It is installable - by mere mortals such as myself?...

I have tried multiple releases of OS4 kernel, each time though I end up having a hard trap right after PMSHELL is initialized. The 201 SMP AOS kernel on the other hand sails right through this w/o any problems.

Initially when I was trying to get this working an individual on our forum (maybe  Valery?) offered to help and troubleshoot further, but it required me to setup a kernel debug environment (com port connection to another machine) so we could capture what's happening during boot. I unfortunately never got around to it, no handy machines to get this configured, etc, etc, subsequently I have not been able to go any further with it.

Some comments though:

1) DOCUMENTATION - early releases were terrible, ad-hoc TXT files here and there, the situation is much today...BUT...there are still multiple modules which although available for d/l one really has no idea if you should try deploying them or not...for example various AMD/SNAP/SDD drivers, DLLs and patches...hard to tell what these are for, and so difficult for someone like myself to attempt to debug this further on the single machine that I have to provide the developers better feedback

2) FUNCTIONALITY - the overall solution seems to be more of a THIS or THAT...meaning, go OS4 all the way, or stay OS2, they were previously providing patches for ACPI < 3.23.01 (I think) but nothing is available for anything newer, so whatever the benefits may be of the new kernel, one does not appear to be able to use them with the officially recognized ACPI support modules

I hope they keep on improving the product, it has real promise and seems to have addressed some key pain-points (heck, the ability to run VBox with VT-x / AMD-V enabled is huge in my opinion).
Title: Re: VirtualAddressLimit
Post by: Dave Yeo on April 15, 2018, 01:13:54 am
I have been following the posting on this thread, and the VirtualBox problems - especially the parts about starting up vbox and settings in CONFIG.SYS

Yesterday I set VIRTUALADDRESSLIMIT=2048, installed and started VBOX multiple times. I tried to create a new virtual machine using a *.VHD from VirtualPC. Which was not successful. But I successfully started VirtualBox.exe multiple times, and attempted to start the new VM I created multiple times.

This morning when I booted up I received the error

Unable to allocate memory for mode info cache!
Fatal error in driver. Use Alt-Ctrl-Del to reboot.

I finally resoved this by changing my VAL back to its original setting, set VIRTUALADDRESSLIMIT=1536

I could then boot up the machine, start VirtualBox.exe, and create a new virtual machine using a Windows XP CD.

I also experience an error after VirtualBox.exe is running, when I start the VM. If I wait the 15 seconds suggested by Mark (I believe) the VM will start.

I am running ARCAOS 5.02. SNAP with dual head.

Hi Doug. How much memory does your video card have? Possibly that is why you need such a low virtualaddresslimit setting. IIRC, there is a way to lower usage, I think Dariusz has experience in that.
Title: Re: VirtualAddressLimit
Post by: Dariusz Piatkowski on April 15, 2018, 01:47:14 am
Doug,

Hi Doug. How much memory does your video card have? Possibly that is why you need such a low virtualaddresslimit setting. IIRC, there is a way to lower usage, I think Dariusz has experience in that.

...follow-up to Dave's post above.

I am using SNAP, my ATI X850 PE card has 256MB on-board memory, but I have this lowered to just 32MB to free up the remaining memory, otherwise SNAP will grab that from the available OS/2 pool.

If you are also using SNAP use the following command @ CLI: 'gaoption vidmem 32'. To confirm that this is active issue the following command 'gaoption show', which will show you all SNAP options, you are specifically looking for the following output (in-between ==> x <== below):

Code: [Select]
Options for ATI Radeon X850 Series (device 0):

  Invert .................. Off
  Rotation ................ Off
  Flipped ................. Off
  Reduced DVI Timings...... Off
  Prefer 16 bit per pixel.. On
  Prefer 32 bit per pixel.. On
  Compressed Framebuffer... On
  Allow DDC BIOS........... On
  PCI bus mastering........ On
  Video memory packets..... On
  Hardware acceleration.... Full
  Multi Head Display....... Off
  VESA DPVL Mode........... Off

Global options for all devices:

  Force VBE Fallback ...... Off
  Force VGA Fallback ...... Off
  Allow non-certified ..... On
  Disable write combining . Off
  Use BIOS for LCD panel... Auto
==>  Video Memory Limit....... 32 Mb <===
  Shared AGP memory size... 4096 Kb
  Use system memory driver. Off
  Disable DDC detection.... Off
  Enable AGP FastWrite..... Off
  Maximum AGP data rate.... 8X
  Virtual Display.......... Off

I am running VIRTUALADDRESSLIMIT=2560 on my machine, 3072 boots as well, but I have to seriously downsize the HPFS386 cache, which just doesn't make sense for me.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 15, 2018, 01:48:46 am
2Doug Clark: VAL=1536 or 2048 may be too small for running VBox! You need VAL=3072. Please reread what has been written in this thread (and another one VBox thread) carefully. You don't need to switch off AMD-V/VT-x in BIOS settings -- only in VBox settings! Your system supports VT-x/AMD-V, but your kernel is not. That's because it is enabled on your system. My system does not support AMD-V, because it is an early Athlon 64 system, so it is turned off by default on my system. So please, turn the checkbox off.

You'd better not start old VPC VM's in VBox, better create a new one, because it has different hardware emulated.

> Unable to allocate memory for mode info cache!
> Fatal error in driver. Use Alt-Ctrl-Del to reboot.

Never seen this. Looks like you have insufficient kernel memory. Are you sure you have correct VAL and decreased JFS cache and THREADS/PROCESSES accordingly? Do you have EARLYMEMINIT=TRUE in CONFIG.SYS BTW?

> I also experience an error after VirtualBox.exe is running, when I start the VM. If I wait the 15 seconds suggested by Mark (I believe) the VM will start.

Why do you need this? You have -1018 error too?

> Does networking work in Vbox?

It works, via NAT, and via OS2TAP driver (but the latter is broken). NAT should work ok. Yes, probably, you have some network configuration error in WinXP. If your OS/2 network connection works, then it should woork in VBox too.
Title: Re: VirtualBox v5.0.51 Testing
Post by: R.M. Klippstein on April 15, 2018, 02:20:23 am
Hi Doug, Here's my 25 cents worth fwiw, I've been using OS4 kernel for close to two years with next to no problems, entirely on AMD Multi-Processor systems with both Panorama & Snap. I've screwed up setting it up a couple of times but once its working seems to be trouble free.

klipp
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 15, 2018, 02:24:56 am
2Dave Yeo: Yes, good note! Modern videocards have too much on-board RAM. Mine, for example has 256 MB (it is Radeon 9600 XT), so I decreased memory mapping to 16 MB in SNAP, which is sufficient for 2D acceleration and my FullHD screen to work. This will greatly economize the kernel address space. The key for VBox to work is to get more available address space to userland (hence, increase VAL to 3072). But increasing VAL requires decreasing JFS cache, if it is too much, and THREADS (to e.g., 511)/PROCESSES (to e.g., 128). Also, you need to control how much free space remained in the kernel arena, with Theseus (see the end of System->Kernel Information->Kernel Memory usage report):

Quote
allocated committed   present  resident  Totals
 --------  --------  --------  --------
 38133000  12D93000  12D15000  111E3000  Total (in bytes)
   918732    308812    308308    280460  Total (in Kbytes)
  897.199   301.574   301.082   273.887  Total (in Mbytes)
< End of THESEUS4 (v 4.001.00) output @ 3:01:58 on 15/4/2018 >

You can see that here is 897.199 of 1024 MB is occupied. So, 100+ MB is free, which is sufficient. I have VAL=3072, PROCESSES=156, THREADS=511, and JFS cache set to 128 MB explicitly:

VIRTUALADDRESSLIMIT=3072
EARLYMEMINIT=TRUE
DLLBASING=OFF
PROCESSES=156
THREADS=511
ifs=d:\os2\jfs.ifs /autocheck:* /cache:128000 /lazy:8,60,5 /p1

PS: Also, you can try to decrease Video memory used  by your video card in BIOS, if you have such setting (in case you have an integrated video card).

2Dariusz Piatkowski: Documentation is good enough, at least, most needed configuration options are described ok. You can ask develpers, if in troubles, they are on this forum and on IRC.

Installation is simple, as any other OS/2 kernel. You have to copy kernel and os2ldr instead IBM's ones, or install QSINIT, which is better to use. Both QSINIT and OS/4 os2ldr support multibooting several kernels. (But there is a number of drivers supplied with the kernel, which should be copied to \os2\boot. These drivers have no name conflicts with existing drivers. AlexT wrote a short HOWTO for quick setup, which is included in latest kernels. Also, there is doscalls1.dll which is optional. It is compatible with IBM's version). Also, there is a multibooting feature of OS/4 os2ldr (and QSINIT works the same time, I use it). So, you can boot multiple kernels from the os2ldr menu (it is better described in QSINIT documentation). Also, OS/4 has built-in CONFIG.SYS editor (activated with Alt-E in loader menu), which is handy if you need to REM out or add something temporarily. It has a good logging feature (a central system log, available for reading via a COM port, or via "copy KERNLOG$ kernlog.txt" after the system booted successfully). Very handy for developers, to debug drivers. Also, there are some enhancements, like hi-res HPET timer support (no hi-res timer bug in apps like Mozilla, like ACPI.PSD has), support for CPU rendez-vous (running the code on all CPU's in parallel), which is used in VBox for VT-x/AMD-V support etc., etc.

Regarding ACPI, yes, OS/4 devs got boring patching each time every new version of ACPI.PSD, so they gave up. But they created their own ACPI module (which is not yet released, it is available for testers, and those who is not too lazy to debug it to make it working, with devs' help). So, ACPI.PSD should be not a problem too.
Title: Re: VirtualBox v5.0.51 Testing
Post by: David Graser on April 15, 2018, 08:44:13 am
Hi Doug, Here's my 25 cents worth fwiw, I've been using OS4 kernel for close to two years with next to no problems, entirely on AMD Multi-Processor systems with both Panorama & Snap. I've screwed up setting it up a couple of times but once its working seems to be trouble free.

klipp

This may seem like a dumb question, but I really don't know which kernel you are referring to when saying OS4 kernel.  Is it the Warp 4 kernel, the eCS kernel, or the kernel that comes with ArcaOS?
Title: Re: VirtualBox v5.0.51 Testing
Post by: OS4User on April 15, 2018, 09:37:20 am
... I really don't know which kernel you are referring to when saying OS4 kernel...


here is some info https://www.os2world.com/forum/index.php/topic,324.0.html
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 15, 2018, 07:12:17 pm
2Doug!

I have been following the posting on this thread, and the VirtualBox problems - especially the parts about starting up vbox and settings in CONFIG.SYS

Yesterday I set VIRTUALADDRESSLIMIT=2048, installed and started VBOX multiple times. I tried to create a new virtual machine using a *.VHD from VirtualPC. Which was not successful. But I successfully started VirtualBox.exe multiple times, and attempted to start the new VM I created multiple times.

This morning when I booted up I received the error

Unable to allocate memory for mode info cache!
Fatal error in driver. Use Alt-Ctrl-Del to reboot.

I finally resoved this by changing my VAL back to its original setting, set VIRTUALADDRESSLIMIT=1536

I could then boot up the machine, start VirtualBox.exe, and create a new virtual machine using a Windows XP CD.

I also experience an error after VirtualBox.exe is running, when I start the VM. If I wait the 15 seconds suggested by Mark (I believe) the VM will start.

I am running ARCAOS 5.02. SNAP with dual head. I ran the YUM install lines from the readme.os2 from version 5.0.6 and I do have the
SET LIBCX_HIGHMEM=2
in my CONFIG.SYS from reading the other posts, although I do not think I have that experimental LIBC package installed.

I also noticed, which I think Valery has now seen, that the default for new VMs has  VT-x / AMD-V enabled.

I did not see a way to turn off AMD-V in my BIOS.


Heh Doug!

Regarding networking, its does work using NAT, which seems to be the default setting.

 I've been using it to download files from the other machines on my network and also from the internet, as I updated all files to the latest level on Windows XP after I connected. A lot of people don't realize it, but you can still upgrade Windows XP from the windows update website, despite service ending in 2014 - there are a couple of things you have to do within windows, but that does work perfectly once those are done.

If its a new copy of the Windows XP VM, you probably have to ensure that your computer is first recognized on your network. Once done, it operates like any other computer. If you need some tips setting that up, let me know.
 
Regarding LIBCX - Look in ANPM (ARCA's package download manager, to see what version your using - the latest regular is LIBCX is 0.6.0-1. The experimental version referred to is 0.6.1-2 and you must have access to the netlabs-exp experimental repository (can be installed through the "Avaiable choIce on ANPM

M
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 15, 2018, 07:48:22 pm
HI Doug!

Had to leave for a few minutes (busy weekend) so I continuance from my previous post.

Regarding LIBCX Part 2 - SET LIBCX_HIGHMEM=2 does not seem to do any harm of there when LIBCX 0.6.0 is installed, although its found to be required for 0.6.1-2 by various people, including me, for VBOX usage.

Regarding AMD-V - There is no AMD-V setting in BIOS AFAIK - I believe it is a virtual setting, which supports aspects of your hardware through the kernel. It uses those aspects the accelerate operation of the VM though the kernel. Simply go to SETTINGS | SYSTEM in VBOX, with the VM you want to change, go to the ACCELERATION tab and uncheck enable VT-X / AMD-V and thats it.

I've found VBOX in testing to be very stable so far, but, like you, I've potentially forgot many of the set-up items I need to do to get the vanilla versions of these various operating systems in VM going.

BTW, I've had limited time to test in the last few days due to corporate year end at the moment, but tried a couple of things:

I tried the MACHINE | CLONE function on Thursday night. VBOX is a lot more complex and feature rich program than VPC, and this feature creates a clone of the VM highlighted. It seems VBOX has an ID number for each VDI it creates, and trying to install another copy of an OS by simply copying the file can cause errors. It also may cause some problems, if you want to take the VDI and install it on another machine.

MACHINE | CLONE brings up two consecutive windows (see attached) to create the clone - creating a full clone seems to be the correct setting to get a VDI you can move around to other machines.

Running that on my upgraded Windows XP VM spiked the CPU usage to about 95% and took about 1 hour and 45 minutes to complete. I was running SeaMonkey occassionally during the time period, but sometimes found some functions extremely slow. Haven't tried the cloned VM but it does now appear in the VBOX window, if I want to test it.

Best!

M

Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 15, 2018, 08:27:34 pm
I'm currently waiting for a full copy of Windows 7 to arrive so I can install and test it in a VM, but decided, since I has an upgrade copy of Windows 7 lying around, to see if I could install one of my extra copies of Windows XP into the Windows 7 VM created.

It was interesting, as the Windows XP system started loading, and then exited with the blue screen attached.

Tried this with two separate authentic copies of Windows XP and got exactly the same screen

It seems that each VM is not just an optimum setting environment for the specified OS - it requires the exact OS to be installed. Interesting.

Anyone else ever try this, just for fun, to see what would happen, and get the same result?

M
Title: Re: Networking in Vbox
Post by: Mark Szkolnicki on April 15, 2018, 08:44:23 pm
Heh Doug!


It has been awhile since I setup networking in Windows XP so this is probably operator error on my part.

BTW: VBOX looks pretty good so far - with my very limited usage at this point.

Doug

Yes things have come a long way, Doug. How much we forget.

Remember the old days - Compuserve? Golden Compass? v2.0 then Warp 3 and Warp 4? 14, 28.8 then 56K modems?

A lot of new faces and a lot of new equipment, and the world goes on as the grey at our temples continues to grow ................

M
Title: Re: VirtualBox v5.0.51 Testing
Post by: ivan on April 16, 2018, 01:28:23 am
Hi Mark,

Are you trying to setup an OS/2 vbox with win 7 running in it and then trying to setup a vbox instance in that win 7 to run XP?  I don't think that will work and is the reason for message you show.

As far as I know vbox will only run one guest OS at a time although it should be possible to run several vbox instances at once on the host machine.  Note I said should because I'm not sure we can with OS/2 as the host machine.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 16, 2018, 04:11:57 am
Hi Mark,

Are you trying to setup an OS/2 vbox with win 7 running in it and then trying to setup a vbox instance in that win 7 to run XP?  I don't think that will work and is the reason for message you show.

As far as I know vbox will only run one guest OS at a time although it should be possible to run several vbox instances at once on the host machine.  Note I said should because I'm not sure we can with OS/2 as the host machine.

Hi Ivan!

That was not the intention.

I was trying to install a Windows 7 update, that I happened to have available, given that I'm still waiting for a full version of Windows 7 to arrive,

The update had to be installed over an existing previous version of Windows. I thought I'd try to install Windows XP in the Windows 7 VM, then install the Windows 7 Update version over it. Obviously that did not work, but if you don't try, you never learn.

Hope that makes sense and Best of the rest of today to you!

M
Title: Re: VirtualBox v5.0.51 Testing
Post by: David Graser on April 16, 2018, 04:52:27 am
... I really don't know which kernel you are referring to when saying OS4 kernel...


here is some info https://www.os2world.com/forum/index.php/topic,324.0.html

I must say that I am impressed with this kernel.  The system is so much more responsive.  Although Firefox took the same amount of time to load as the IBM kernel, every subsequent startup was fast.  The CPU does not appear to trash about like with the IBM kernel. Open Office appears to load faster and my CPU is running cooler.  My video is faster also.  So far, I have had no problems with the very latest release.   Thanks OS4user for the heads up.

David
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 16, 2018, 04:44:47 pm
Good Morning All!

This week I'm planning to start testing use of various programs in a VBOX VM, to see how things work out.

A lot of my consultancy involves technical writing, so I spend a lot of time on programs like Word and graphics design programs like CorelDraw, so will be using those a lot in the next few days.

I can say that they do open (a hopeful start) and that I can access files from the host computer and others, through my peer network. I did try enabling the "shared clipboard" in VBOX through SETTINGS | GENERAL then the Advanced Tab and found that attempting to cut and paste between the host and the VM locked things up. Only tried "Bidirectional" at the moment (also has a Host to VM or VM to Host option), but, I expect, this being a developmental version that some code linking the OS and the VM clipboards is still needed.

I also decided to install the new version of SeaMonkey that Dave made available (2.42.9-exp3), as I was running into some memory problems with 2.35. The new 2.42.9 is actually the first drop of that version that I could get to work and it seems to be a little faster and more responsive (2.35 would take about 20 seconds to load and sometimes I would get pauses at the beginning of a session or when loading a site, and also the occasional random temporary lock-up with 2.35). The problems with 2.35 were relatively minor in the scheme of things, but this seems to be an improvement, so plan to stick with 2.42.9

Memory usage for SeaMonkey has skyrocketed though - reading about 437 Mb being used, where 2.35 was using about 330 Mb, just looking at SysInfo - I haven't tested it with Theseus, now installed, or with the app Dariusz provided. Will have to try starting up the VM again at some point with both running to see what the memory is doing.

Best of the start of the week!

M

Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 16, 2018, 04:50:05 pm
[
... I really don't know which kernel you are referring to when saying OS4 kernel...


here is some info https://www.os2world.com/forum/index.php/topic,324.0.html

I must say that I am impressed with this kernel.  The system is so much more responsive.  Although Firefox took the same amount of time to load as the IBM kernel, every subsequent startup was fast.  The CPU does not appear to trash about like with the IBM kernel. Open Office appears to load faster and my CPU is running cooler.  My video is faster also.  So far, I have had no problems with the very latest release.   Thanks OS4user for the heads up.

David

Heh David!

Was just curious.

What OS and what version have you installed the OS/4 kernel on?

I have boxes running eCS and ArcaOS in various configurations. Thought I might try the latest version on one of my experimental units.

Best!

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: David Graser on April 16, 2018, 06:27:26 pm
Hi Mark

I am using Arcaos 5.02.

This kernel is recent after a long break of no work.

http://ru2.halfos.ru/core/downloads/kernel/os4/os2krnlSVN5087_unoff.zip

One has to be careful setting it up, but if you make a mistake it allows one to use the original boot loader which boots the original kernel to fix whatever mistake one makes setting it up.



Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 16, 2018, 06:37:46 pm
Hi David!

Hi Mark

I am using Arcaos 5.02.

This kernel is recent after a long break of no work.

http://ru2.halfos.ru/core/downloads/kernel/os4/os2krnlSVN5087_unoff.zip

One has to be careful setting it up, but if you make a mistake it allows one to use the original boot loader which boots the original kernel to fix whatever mistake one makes setting it up.


Do you use QSINIT? I have tested it in the past in passing and was wondering if it might help in the setup (ie. allow different kernels to boot?

M
Title: Re: VirtualBox v5.0.51 Testing
Post by: David Graser on April 16, 2018, 06:44:28 pm
Hi David!

Hi Mark

I am using Arcaos 5.02.

This kernel is recent after a long break of no work.

http://ru2.halfos.ru/core/downloads/kernel/os4/os2krnlSVN5087_unoff.zip

One has to be careful setting it up, but if you make a mistake it allows one to use the original boot loader which boots the original kernel to fix whatever mistake one makes setting it up.


Do you use QSINIT? I have tested it in the past in passing and was wondering if it might help in the setup (ie. allow different kernels to boot?

M

I don't use QSINIT.  I believe the OS4 kernel comes with its own boot loader which is setup with an ini file.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 17, 2018, 03:10:36 am
2David Graser: QSINIT is mostly compatible with os2ldr from OS/4, but it has more features
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 17, 2018, 03:29:33 am
2Mark Szkolnicki: Updating WinXP to win7 is possible, but note that win7 will require significantly more resources than winxp. The bare minimum for Win7 is 768 MB of RAM, but in case you disabled all "bells and whistles" like Aero and visual effects, and enable VT-x/AMD-V. Otherwize, it will run too slow. WinXP with 512 MB of RAM is an optimal variant even without HW virtualization. So, I'd not recommend to upgrade it to Win7.

> I can say that they do open (a hopeful start) and that I can access files from the host computer and others, through my peer network. I did try enabling the "shared clipboard" in VBOX through SETTINGS | GENERAL then the Advanced Tab and found that attempting to cut and paste between the host and the VM locked things up. Only tried "Bidirectional" at the moment (also has a Host to VM or VM to Host option), but, I expect, this being a developmental version that some code linking the OS and the VM clipboards is still needed.

Shared Folders should work with different guests (excluding OS/2 atm. I did not finished the shared folders IFS yet. But it currently allows navigating the file system and viewing the files. Though, it can crash the VM easily, so it is not stable yet, but partially working). So, host support for shared folders on OS/2 is working, but guest support is not yet.

Inversely, Shared clipboard host support is not yet finished, so it can easily hang your host system! So, beware. Support for shared folders on OS/2 guests should work, though.

2ivan: running two different VM's on top of OS/2 host is possible. At least, I ran two VM's, one with WinXP, another with Knoppix LiveCD (Linux). They are not prevent to run each other. Running a VBox inside VBox is not poossible, AFAIK, because VBox does not support nested virtualization. QEMU supports it. So, I was able to run VBox inside QEMU. It worked, but very slow. KolibriOS started for an hour, or so, in such VBox instance. (This is without HW virtualization/KVM. With KVM enabled, VBox runs relatively well inside QEMU).
Title: Re: VirtualBox v5.0.51 Testing
Post by: David Graser on April 17, 2018, 07:20:24 am
Hi Mark

I am using Arcaos 5.02.

This kernel is recent after a long break of no work.

http://ru2.halfos.ru/core/downloads/kernel/os4/os2krnlSVN5087_unoff.zip

One has to be careful setting it up, but if you make a mistake it allows one to use the original boot loader which boots the original kernel to fix whatever mistake one makes setting it up.

This is the latest that corrected a problem I was encountering at startup.

http://www.gus.biysk.ru/os4/os2krnlSVN5094_unoff.zip

As was mentioned earlier, once setup correctly, it works great. 
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 17, 2018, 07:29:34 pm
2Mark Szkolnicki: Updating WinXP to win7 is possible, but note that win7 will require significantly more resources than winxp. The bare minimum for Win7 is 768 MB of RAM, but in case you disabled all "bells and whistles" like Aero and visual effects, and enable VT-x/AMD-V. Otherwize, it will run too slow. WinXP with 512 MB of RAM is an optimal variant even without HW virtualization. So, I'd not recommend to upgrade it to Win7.

> I can say that they do open (a hopeful start) and that I can access files from the host computer and others, through my peer network. I did try enabling the "shared clipboard" in VBOX through SETTINGS | GENERAL then the Advanced Tab and found that attempting to cut and paste between the host and the VM locked things up. Only tried "Bidirectional" at the moment (also has a Host to VM or VM to Host option), but, I expect, this being a developmental version that some code linking the OS and the VM clipboards is still needed.

Shared Folders should work with different guests (excluding OS/2 atm. I did not finished the shared folders IFS yet. But it currently allows navigating the file system and viewing the files. Though, it can crash the VM easily, so it is not stable yet, but partially working). So, host support for shared folders on OS/2 is working, but guest support is not yet.

Inversely, Shared clipboard host support is not yet finished, so it can easily hang your host system! So, beware. Support for shared folders on OS/2 guests should work, though.


Good Morning Valery!

By updating a Windows XP installation to Windows 7 has that been done within a VM originally created for Windows XP, or trying to install it into a VBOX VM created specifically for Windows 7 originally (ie. one using NEW | Create New Virtual Machine and choosing "version Windows 7)?

Thats what I tried, just to see if it would work, as I have no full OS copy of Windows 7 at the moment (just an upgrade copy I had lying around).

I will still try to create a VM installation of Windows 7 on this machine, just for testing purposes, but I suspect, given the memory limitations you describe Valery, your probably right.

Having used the Windows XP VM now for a day with Word, and a Quickbooks accounting package installed, the VM is extremely useable, running with now slowdowns I could decern, and in fact seems faster than the VPC installation I use of Windows XP.

Shared folders does indeed work. I was also able to install a printer driver for an HP3500 Colour LaserJet and print out from the VM. I didn't see a reason why it would not, but I have the printer installed on the network, through a print server installed at IP 192.168.1.55, so the installation is a little trickier than installing a local printer. I may try to test this by installing a local printer in future, as I do have an HP2100 LaserJet gathering dust, as I've been having trouble installing it to the printer server, as its connection id through a LPT port, and it doesn't seem to be recognized through a LPT to USN connection on the print server.

Thanks for confirming about shared clipboard. I don't know if you are also planning to do something with drag and drop in future as well, but it would be useful, for some of the items I do, in my consiltancy.

But very good work, so far, especially for a development version, Valery. I will continue to test and report findings as I go along!

A fine day to you (and everyone else) and Best!

M

Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 18, 2018, 04:40:47 am
2Mark Szkolnicki: Yes, I'm still working on a host-side shared clipboard DLL and guest-side shared folders IFS for OS/2. No work is done on drag-n-drop, yet. Actually, I have no idea how should it work, and what could be missing. It seems, I was able to drag files from OS/2 desktop to VM desktop, for WinXP guest (or vice versa). But I'm not sure how should it work. So, if you have an idea, what's not working, please share.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on April 18, 2018, 11:47:37 pm
Dariusz:  Thank you for the instructions on how to limit the memory SNAP consumes. The ATI video card I have has 256 MB. I read the discussion on the SNAP topics but didn't see how to actually effect the setting. I followed your advice and set to 32MB.

I have set the VirtualAddressLimit = 3072 and the jfs .... cache=64000

Vbox comes up fine - except for the required 15 second wait between when the VirtualBox Manager screen appears and I can start a VM; wait 15 seconds VM starts fine.

NAT networking appears to be working OK. Pings do not work, and I could not activate Windows XP which lead me to believe networking wasn't working. But IE opens MSN.NET the HP printer driver for

I installed the HP printer driver for HP452 and it prints a test page.

The Adobe universal windows postscript driver installer v 1.06 starts but never completes. I have tried multiple times to run the installer, and once got the splash screen to display but nothing further. It is an InstallShield installer.

Thanks guys for all your help.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on April 19, 2018, 07:36:03 pm
The installer for FrameMaker also starts but then hangs. That is also InstallShield. I suspect that part of it is 16 bit. This is all with the OS2 kernel.

When the install hangs, and I reboot, Windows XP complains about WOWEXEC - saying it can't stop WOWEXEC, which is why I believe it is related to the 16 bit part of the installer.

But it appears that if I run a 16-bit program first (I am using NOTEPAD.EXE (from C:\OS2\MDOS\WINOS2), the installer is successful more often. Not all the time, but enough to install FrameMaker, Distiller, and ATM, all of which had problems. I also seen to have better success if I copy the files to a drive local to the VM before starting the install.

Now that all that is installed, vBox is very impressive. The mouse pointer is strange, sometimes laging slightly behind the mouse movement, but the speed and performance are impressive - without hardware virtualization.
Title: start up error
Post by: Doug Clark on April 19, 2018, 11:54:14 pm
I have twice now gotten a error trying to startup a VM. The error is

PIIX3 cannot attach to the primary slave (verr_io_not_ready)

I solve this by removing the optical drive in the settings for the VM, and then start the VM. After the VM has been rebooted and shut down I can reattach the optical drive.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on April 20, 2018, 03:30:32 pm
I cannnot

THREADS=511
PROCESSES=156

in CONFIG.SYS on my machine. When I do the Netquestion  search engine doesn't start. Which doesn't make sense but that is what happens. Netquestion is the help system for the later IBM C/C++ compilers and DB2 clients.

But it doesn't appear that I need to because vBox runs without those. (I do have VirtualAddressLimit=3072 and I have reduced the memory used by SNAP.)
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 24, 2018, 02:30:43 am
So, increase THREADS to say, 612, or 712. This depends on how much threads Netquestion needs. Usually I don't need so much threads.

> PIIX3 cannot attach to the primary slave (verr_io_not_ready)

Is this a physical CD? Probably, you extracted the CD from the drive, then that error is normal.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on April 24, 2018, 05:58:50 am
Valery: yes it was a physical CD drive. I don't remember if there was a disk in the drive at the time, and will note that the next time I see the error so that I can report one way or the other.

As for the memory. How much does vBox need? What figure am I looking for, in either Theseus or shmemmon?  That might be easier than to shoot for specific numbers on config.sys parameters. (Although listing the config.sys parameters and the related discussion was very helpful for me in adjusting memory. ) For example I have an SSD drive and a spinning drive with a large built in cache, so specifying a JFS cache of 64 MB doesn't seem to degrade performance at all. I could probably go even lower without noticing a slow down.

I know you posted before about Theseus: to use Kernel Memory Usage and scroll to the bottom. But I didn't understand which figure (Allocated, committed, present, resident) I should look at and what the number should be.

Finally - I hope all the discussion about issues and problems with vBox doesn't drown out the fact that this version is pretty amazing. I was very pleasantly surprised at how well vbox works and how much is there.
Title: Re: VirtualBox v5.0.51 Testing
Post by: roberto on April 24, 2018, 08:21:04 am
When updating the new anpm packages, the vbox no longer works and I can not install the previous versions.

Fallo al crear el objeto COM de VirtualBox.
La aplicación se cerrará ahora.
Receptor RC: NS_ERROR_ABORT (0x80004004)
Title: Re: VirtualBox v5.0.51 Testing
Post by: hannos on April 24, 2018, 09:20:05 am
The problem is libcx 0.6.2-1: Go back to version 0.6.0-1 and vbox will work again.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 24, 2018, 10:32:54 am
2Doug Clark: You don't understand the word "allocated"? So, it your VAL=3072, then you have 1 GB of kernel virtual memory. Then 800 MB allocated means that 800 MB of kernel virtual memory is occupied, so 200 MB is free. So, you should orient on this number. I don't know how much memory VBox needs -- it depends on VM sizes. If you'll allocate 512 MB for a VM, then about 600 MB will be occupied. But for itself, VBox occupies not so much memory.

2roberto: Did you read the whole thread? If so, you should notice that such an error can occur in case you have newer libcx version installed. In this case, you need to try SET LIBCX_HIGHMEM=2.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 24, 2018, 09:25:06 pm
Valery: yes it was a physical CD drive. I don't remember if there was a disk in the drive at the time, and will note that the next time I see the error so that I can report one way or the other.

As for the memory. How much does vBox need? What figure am I looking for, in either Theseus or shmemmon?  That might be easier than to shoot for specific numbers on config.sys parameters. (Although listing the config.sys parameters and the related discussion was very helpful for me in adjusting memory. ) For example I have an SSD drive and a spinning drive with a large built in cache, so specifying a JFS cache of 64 MB doesn't seem to degrade performance at all. I could probably go even lower without noticing a slow down.

I know you posted before about Theseus: to use Kernel Memory Usage and scroll to the bottom. But I didn't understand which figure (Allocated, committed, present, resident) I should look at and what the number should be.

Finally - I hope all the discussion about issues and problems with vBox doesn't drown out the fact that this version is pretty amazing. I was very pleasantly surprised at how well vbox works and how much is there.

2DougClark:

Hi Doug!

I think you got the solution from Valery already, but I got that CD error when I attached a physical drive inside the VM using DEVICES | OPTICAL DRIVES | HOST DRIVE (in my case either S: or T:) and did not after using it click "Remove Drive from Virtual Disk", at the bottom of that same menu.

If you shut down the VM before doing that, and remove the disc, you will get the PIIX3 error. Putting any disk back in the physical CD will make the error disappear.

I did try once to attempt to attach or detach an Optical drive from the VirtualBox main window, that comes up first, through choosing a VM, then clicking SETTINGS | STORAGE , then in the Storage tree window choosing an empty slot and going over to attributes. Clicking on the Optical Drive Icon there, I was also able to attach a host drive or an image.

Also I would like to echo your comments about this version of VBox. I've used it now for approximately a week doing word processing, graphics design and using a QuickBooks accounting program (actually a Windows 3.1 program but works perfectly in Windows XP) and things have worked flawlessly, at speeds which are indistinguishable from the base operating system (in this case ArcaOS, but plan to test on eCS 2.1 and 2.2 as well.

It is one of the reasons I started this tread in the first place, as "Testing" is really what we're all doing, and I hope others will also post what they found has went well, as well as the problems, as we learn together. Having taken a crack at reading the VirtualBox manual, it is a tremendously feature rich, complex program in comparison to VirtualPC. I want to encourage the developers in their efforts, as this, as it develops, will be an extremely useful tool for the community.

Best!

M
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 24, 2018, 09:34:27 pm
The installer for FrameMaker also starts but then hangs. That is also InstallShield. I suspect that part of it is 16 bit. This is all with the OS2 kernel.

When the install hangs, and I reboot, Windows XP complains about WOWEXEC - saying it can't stop WOWEXEC, which is why I believe it is related to the 16 bit part of the installer.

But it appears that if I run a 16-bit program first (I am using NOTEPAD.EXE (from C:\OS2\MDOS\WINOS2), the installer is successful more often. Not all the time, but enough to install FrameMaker, Distiller, and ATM, all of which had problems. I also seen to have better success if I copy the files to a drive local to the VM before starting the install.

Now that all that is installed, vBox is very impressive. The mouse pointer is strange, sometimes laging slightly behind the mouse movement, but the speed and performance are impressive - without hardware virtualization.

2DougClark

Hi Doug!

I was curious whether you finally activated your copy of the WindowsXP VM, and then upgraded the base XP operating system with service pack 3 and all the Windows updates which were released up to 2014?

I have all the updates installed, but don't use FrameMaker so can't  test install myself. But I do know that some programs need various of the Windows XP Updates installed from their site, to function properly with InstallShield.

It might be your problem, or it may not be, but, as always it never hurts to ask the stupid questions.

Best!

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 24, 2018, 10:01:45 pm
Hi All!

I had a couple of minutes this afternoon to post.

I've been using VirtualBox, 5.0.51, this development version, now for about a week continuously, and have found, for something in development, it has been extremely stable. I have found no reason, for the moment, to switch back to the Virtual PC v5.1 version I also have installed.

From an additional testing standpoint, I have been able to create a snapshot of the running Windows XP VM.

According to the VBOX manual:

"Snapshots" is a feature of "Oracle VM VirtualBox" that allows you to save and then revert back to the exact state of a virtual machine. At any point in time for any virtual machine, you can take a "snapshot".

It took approximately 40 minutes on my machine and pegged the CPU usage (I have four AMD cores on this machine but did not check to see what was happening at the time with the various CPU's). Usage of other programs was extremely slow as a result, while creating the snapshot. If I have a few moments, I will attempt to try that again, examining whats happening with both memory and CPU usage.

I also was able to "save machine state", on shutdown. Similar to VirtualPC, doing that allows you to resume usage of the VM in the exact state you left it. Closing using FILE | CLOSE then choosing "Save Machine State" took approximately 40 seconds to save the VM.

When I started up again, later, I ran into the same memory related problems as with a VM that was in total shutdown, getting the "-1018 error", unless I waited about 15 seconds before opening the VM. But the saved machine state opened up, again taking about 40 seconds to load, and I was able to resume using the VM session where I left off.

One thing I haven't tried as of yet, is to see whether I can mount a USB stick to transfer data into the VM. Valery, if you're around, is that ability enabled, or  or is there still code to write, for that to work?

Its been a very pleasant testing experience so far, and Best!

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 24, 2018, 10:33:00 pm
2Mark Szkolnicki:

> One thing I haven't tried as of yet, is to see whether I can mount a USB stick to transfer data into the VM. Valery, if you're around, is that ability enabled, or  or is there still code to write, for that to work?

It requires the USB passthrough feature, which is not yet finished. But can mount the stick under host OS/2 and when the drive letter appears, try to start the VM with a .vmdk file added for that drive letter.

vmdk files can be created with VboxManage.exe like this:

Quote
VBoxManage internalcommands createrawvmdk -filename 1.vmdk" -rawdisk 1:

Where 1: is a 1st harddisk, and if you'll add the 1.vmdk file to your Media menu, VBox will access all physical 1st hard disk. If you'll create the x.vmdk file like this:

Quote
VBoxManage internalcommands createrawvmdk -filename x.vmdk" -rawdisk \\.\x:

And then add the resulted x.vmdk to Media menu, then only host drive letter x: will be attached to the VM. This way, you can theoretically add any drive, including the flash stick. But this disk should be ready when VM starts, otherwise VM will fail to start.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on April 25, 2018, 05:35:07 am
Valery: thanks for the explanation on Theseus. I am still confused on how a virutalAddressLimit=3072 = 1 GB kernel memory, but I will take your word for it.

Mark: I never could get WindowsXP to activate through the internet - it kept failing saying it couldn't reach the server or something like that. But I did get the copy activated through Microsoft's dial up number - which is completely automated. You enter a cell phone number, it sends a text link to that number, on that link/web page you input the product id, and if Microsoft is happy they send you the activation code.  Their system would not send an activation number for an install off of a Dell CD, but did send an activatation for an install off of a different CD.

I then installed Service pack 3. I have not installed any other Windows updates. The version of FrameMaker I installed (V 5.5) was released before XP was available, so the problems could be related to some update I should have installed for XP.  Or it could be that InstallShield sucks. The Adobe Postscript Driver was also an InstallShield and it also  was flakey. However the 16 bit programs I have run (Notepad and Acrobat Distiller 3)  run fine.

So what I have installed and running are: FrameMaker 5.5, Acrobat Reader v 7, Larsen Commander which are all 32 bit, and  Arcobat Distiller v 3, which is 16 bit.  When the USB passthrough gets working I will be able to sync my Palm TX to the desktop running in XP (I am probably the last person on the planet still using a palm pilot.)

All these run very well, although the mouse pointer is a little funky: it seems to lag slightly behind the movement of the mouse.

I could not get Virtual PC 5.1 to run on my machine so Vbox is the only alternative - well the only one I have tried. And it is a very good alternative
Title: Re: VirtualBox v5.0.51 Testing
Post by: OS4User on April 25, 2018, 01:40:14 pm
I am still confused on how a virutalAddressLimit=3072 = 1 GB kernel memory, but I will take your word for it.

Address space (4Gb) is divided into several parts:
Conventional Memory                  :  0 - 512Mb
High Memory Arena                     :  512Mb - VAL
System Arena (kernel memory)   :  VAL - 4Gb

(VAL stands for Virtual Address Limit)

Note that this division does not have anything in common with availability of phys mem.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 25, 2018, 03:25:17 pm
2 Doug Clark: OS/2 has a 4 GB address space. If VAL=3 GB, then 4-3=1 GB remains in the kernel arena, which is above 3 GB.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on April 25, 2018, 03:58:03 pm
Thanks OS4 for the mem-map. That explains a lot. And thanks Valery for the explanation.

It appears that VAL is the dividing line between kernel memory at the top, and high memory below that. The higher the VAL the less kernel memory and more high memory; the lower the VAL the more kernel memory and the less high memory.

So it looks like you are really trying to maximize the amount of high memory (by setting VAL to a higher number), while making sure the stuff that has to be allocated (or reside in) kernel memory can still fit. Stuff like the JFS cache and the structures for threads and processes.

If I am understanding this correctly when Vbox and Firefox will not run at the same time, the issue is not enough high memory and the solution is to set VAL higher.

If my system is unstable (without Firefox or Vbox running) the issue is not enough kernel memory, and the solution is to lower VAL or reduce the demand for kernel memory by adjusting JFS cache and processes/threads lower.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 25, 2018, 04:33:49 pm

Mark: I never could get WindowsXP to activate through the internet - it kept failing saying it couldn't reach the server or something like that. But I did get the copy activated through Microsoft's dial up number - which is completely automated. You enter a cell phone number, it sends a text link to that number, on that link/web page you input the product id, and if Microsoft is happy they send you the activation code.  Their system would not send an activation number for an install off of a Dell CD, but did send an activatation for an install off of a different CD.

I then installed Service pack 3. I have not installed any other Windows updates. The version of FrameMaker I installed (V 5.5) was released before XP was available, so the problems could be related to some update I should have installed for XP.  Or it could be that InstallShield sucks. The Adobe Postscript Driver was also an InstallShield and it also  was flakey. However the 16 bit programs I have run (Notepad and Acrobat Distiller 3)  run fine.

So what I have installed and running are: FrameMaker 5.5, Acrobat Reader v 7, Larsen Commander which are all 32 bit, and  Arcobat Distiller v 3, which is 16 bit.  When the USB passthrough gets working I will be able to sync my Palm TX to the desktop running in XP (I am probably the last person on the planet still using a palm pilot.)

All these run very well, although the mouse pointer is a little funky: it seems to lag slightly behind the movement of the mouse.

I could not get Virtual PC 5.1 to run on my machine so Vbox is the only alternative - well the only one I have tried. And it is a very good alternative

Heh Doug!

Yes, Dial-up activation still works. One other thing I forgot to mention is that the Windows websites now do not seem to like the IE6 that comes with Windows, probably due to changes in the website components that have occurred over the years. It seems to need IE8 or higher to connect to the Windows Update site, so perhaps the same applies to the Windows XP activation site, if it still exists.

If your interested in updating XP components and getting security updates from Windows Update in future, I found this procedure to work, for Windows XP at least:

1) Install the base Windows XP OS into the VM.

2) Install Windows XP Service Pack 3 (if not already installed with the base system)

3) Install Internet Explorer 8 - the Windows Update site will not work with IE6, which comes with Windows XP originally.

4) Connect to Windows Update

5) Download the latest installation Pack that works for Windows XP (v4.5 - WindowsXP-KB942288-v3-x86). It will choose this automatically and after a reboot ActiveX will work.

6) It will also ask to install two update files (Geniune Advantage validation (KB898461) and Update for Windows KB898461))

7) After the necessary reboots for the various steps above you should have access to all Windows XP updates prior to discontinuing support in March 2014 (including .NET v4, if you need it).

8) A final key security update was issued (KB4012598) was issued for all Windows platforms (including Windows XP), to address security flaws related to the wannacry virus scare. This should be installed as a separate download and runtime)

Its has indeed been many years since I heard the name "Palm Pilot" used anywhere in polite company - but if it ain't broke and works perfectly, don't fix it, Sir  :)

I have also noticed that the mouse pointer is a little bit flakey as well - I hadn't put it down to a delay problem, but I suspect you are right, on sober second thought - a VBOX VM pointer occasionally chooses a menu item above the one I was pointing to.

And regarding Virtual PC v5.1 - since the advent and widespread use of multicores, its been difficult to run on newer machines, and many times attempting to start the program could cause lock-ups (probably both memory and CPU compatibility problems). Putting in the activation key could also lock up the system, so installation was starting to become a little thaumaturgical (Where is Merlin when you need him? <sic>).

It could also sometimes be painfully slow in execution as well with newer machines. I bought 4 licenses, just as Microsoft bought the product, to ensure the product would still be available to me for business use, and it has performed faithful service for many years. But the VBOX alternative has proved fantastic so far.

Best!

M

Title: Re: VirtualBox v5.0.51 Testing
Post by: Pete on April 26, 2018, 12:56:30 am
Hi All

Probably 1 area that vpc works much better for me than vbox is networking as in vpc it is possible to use a static ip address for the guest using the same ip address range as my local network. This is not possible in vbox it seems.

I am currently trying to work out how to connect to my printer from a vbox guest. The vbox guest has a dhcp allocated ip address in the 10.0 range; the printer is at 192.168.1.252. What magic incantation is required to get the vbox guest to be able to print?


Regards

Pete
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 26, 2018, 05:01:57 am
Hi All

Probably 1 area that vpc works much better for me than vbox is networking as in vpc it is possible to use a static ip address for the guest using the same ip address range as my local network. This is not possible in vbox it seems.

I am currently trying to work out how to connect to my printer from a vbox guest. The vbox guest has a dhcp allocated ip address in the 10.0 range; the printer is at 192.168.1.252. What magic incantation is required to get the vbox guest to be able to print?


Regards

Pete


Heh Pete!

Since you want some thaumaturgy tonight. I'll give you my incantation, which has worked on the Windows XP VM.
(Was that why Warp 4 was dubbed "Merlin" BTW?)

What OS VM are you using? And can the VM be seen by your network?
(Setting up the new Windows XP VM I had to make sure it was part of the workgroup, something I hadn't done in a while)

I print out to static address 192.168.1.55 on my network (print server) and am currently printing out to an HP3500 that I've set up there successfully from the VM.

To do that I created a standard TCPIP Port and print out to that (see picture attached).

I actually haven't looked to see what IP VBOX is reporting, but using the default NAT adapter setting, and not setting any particular dynamic or static IP for the VM, I'm connected to PALNET, my peer network, and can access all my computers and the HP3500 from at least the Windows XP VM.

Looking at the moment I see HYPERGHOST, my Windows XP VM is visible to the network through ArcaMapper (Netdrive) (picture attached) - I haven't tried to mount and connect to it, using that method, as of yet.

Hope that little bit of magic helps a bit and Best!

Mark

Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 26, 2018, 03:15:28 pm
2Pete: VBox currently works via NAT (TAP.OS2 driver functionality is currently broken). So VBox guest is seen under host IP (from 192.168.x.x range). No special magic words are needed, I suppose.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 26, 2018, 03:47:04 pm
Hi All

Probably 1 area that vpc works much better for me than vbox is networking as in vpc it is possible to use a static ip address for the guest using the same ip address range as my local network. This is not possible in vbox it seems.

I am currently trying to work out how to connect to my printer from a vbox guest. The vbox guest has a dhcp allocated ip address in the 10.0 range; the printer is at 192.168.1.252. What magic incantation is required to get the vbox guest to be able to print?


Regards

Pete


Good Morning, Pete! (and everyone else)

Being the heretic I am, I (Horror of Horrors) actually READ THE VBOX MANUAL today related to networking, and what I found was most interesting.

What you described, Pete, related to range addressing is actually a designed feature of Network Address Translating (NAT) in VBOX, and a rather elegant one.

According to Section 6.3 of the Manual (quoted):

Network Address Translation (NAT) is the simplest way of accessing an external network from
a virtual machine. Usually, it does not require any configuration on the host network and guest
system. For this reason, it is the default networking mode in VirtualBox.

A virtual machine with NAT enabled acts much like a real computer that connects to the
Internet through a router. The “router”, in this case, is the VirtualBox networking engine, which
maps traffic from and to the virtual machine transparently. In VirtualBox this router is placed
between each virtual machine and the host. This separation maximizes security since by default
virtual machines cannot talk to each other.

The disadvantage of NAT mode is that, much like a private network behind a router, the virtual
machine is invisible and unreachable from the outside internet; you cannot run a server this way
unless you set up port forwarding (described below).

The network frames sent out by the guest operating system are received by VirtualBox’s NAT
engine, which extracts the TCP/IP data and resends it using the host operating system. To an
application on the host, or to another computer on the same network as the host, it looks like
the data was sent by the VirtualBox application on the host, using an IP address belonging to the
host. VirtualBox listens for replies to the packages sent, and repacks and resends them to the
guest machine on its private network.

The virtual machine receives its network address and configuration on the private network
from a DHCP server integrated into VirtualBox. The IP address thus assigned to the virtual
machine is usually on a completely different network than the host. As more than one card of
a virtual machine can be set up to use NAT, the first card is connected to the private network
10.0.2.0, the second card to the network 10.0.3.0 and so on.

In Virtual PC, we had to install the Innotek Virtual Switch between VPC and the host to access the network. It seems NAT does this automatically, with VBOX acting like a VPN, and NAT translating between the Host network and VBOX's internal network.

Reading Section 9.11 of the manual, it indicates that NATs address range can be placed in a different range using VBOXMANAGE, but, given the way it operates, I can't see a reason for us to do that, unless under much more complex circumstances than I think any of us would encounter.

If anyone wanted to use a VBOX VM as a server or for other reasons, directly accessing the internet, there is also a port forwarding function that can be configured, as outlined in Section 6.3.1 of the manual.

The developers of VBOX have put a tremendous amount of work into this puppy, which is not immediately evident unless you delve into its workings. As a technical writer myself (and an author) I'm also impressed by their manual, which is very well and clearly written.

Best to All today!

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 26, 2018, 04:10:34 pm
Hi All!

BTW, I haven't mentioned this before, but during testing, I found the way to enable VBOX Help.

I attempted to access HELP | CONTENTS a few weeks ago, and found nothing there. But it did ask if I would like to download the manual from the VBOX website.

It places the latest User manual available ((UserManual.PDF, in this case v5.2.8) in the Default Machine Folder specified under FILE | PREFERENCES in the main window. The next time you click on HELP | CONTENTS, it will open the PDF in Lucide, at least on ArcaOS 5.0.1.

The current manual, despite being 378 pages long, is very well organized but not meant to be read cover to cover at one sitting. The parts I have consulted are very easy to both read and understand, even for a non programmer like myself.

Best!

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on April 26, 2018, 04:51:21 pm
2Mark Szkolnicki:

> In Virtual PC, we had to install the Innotek Virtual Switch between VPC and the host to access the network. It seems NAT does this automatically, with VBOX acting like a VPN, and NAT translating between the Host network and VBOX's internal network.

In VPC, there is NAT too (aka "Shared networking"), and Virtual Switch appeared not immediately. there were old times, when Virtual switch did not yet existed. There is Virtual switch analogue in VBox too, called TAP.OS2. It acts as a node of a virtual network too. (But it is non-functional, currently). NAT isn't a VPN analogue, it just translates IP addresses to a number of ports on NAT side. Each connection from vietual network is assigned a port on NAT side.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Pete on April 26, 2018, 06:52:43 pm
Hi All

Printer now working in my Win7 guest - not sure what went wrong on the previous 2 attempts...

My other VBox guest is lubuntu 17.10 which helpfully informs me "cups-client-error-not-possible" when I try to setup a printer. I suspect that lubuntu needs some cups updates as I do not see the necessary bjnp cups backend anywhere in the installation. Will report back if/when I resolve this.

@Mark
Setting up a standard tcpip port will not work with Canon printers that use bjnp but thanks for the thought.


Probably Win7 and lubuntu are not the best guests as I find both guests painfully slow to boot and run any application. Hopefully as VBox development continues this will improve.


Regards

Pete
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 26, 2018, 07:05:42 pm
Hi All

Printer now working in my Win7 guest - not sure what went wrong on the previous 2 attempts...

My other VBox guest is lubuntu 17.10 which helpfully informs me "cups-client-error-not-possible" when I try to setup a printer. I suspect that lubuntu needs some cups updates as I do not see the necessary bjnp cups backend anywhere in the installation. Will report back if/when I resolve this.

@Mark
Setting up a standard tcpip port will not work with Canon printers that use bjnp but thanks for the thought.


Probably Win7 and lubuntu are not the best guests as I find both guests painfully slow to boot and run any application. Hopefully as VBox development continues this will improve.


Regards

Pete


Heh Pete!

Glad to see you solved it. How are you connecting out from your guests, port wise, as it sounds like your using network printing as well?

I've found Windows 7 setups for networking to be wonky, as it wants to do everything for you, its way, and assumes you have the IQ of a carrot. Never tried Iumbutu (which I assume is a version of Umbutu, being more exotic OS challenged), so know nothing about how your interacting there.

Almost all my network printing tricks I learned during my war with getting CUPS to work in the past, when I switched from local printers to a Print Server.

Best!

M
Title: Re: VirtualBox v5.0.51 Testing
Post by: Pete on April 26, 2018, 08:21:32 pm
Hi Mark

There is no interaction between VBox guest and rest of network apart from printer access as it is not required. Also, come to think of it, I am using IBM Peer in preference to samba for my local network so it would not work.

As far as lubuntu goes: it is a ubuntu variant and like all "nixes" is a PITA. The final straw is that lubuntu decided to boot to a totally mangled display for no obvious reason and after several reboots with the same result is no longer a VBox guest.

Your comment "I've found Windows 7 setups for networking to be wonky, as it wants to do everything for you, its way, and assumes you have the IQ of a carrot." seems to apply to most nixes as well.


Regards

Pete
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on April 26, 2018, 09:55:06 pm
Hi Mark

There is no interaction between VBox guest and rest of network apart from printer access as it is not required. Also, come to think of it, I am using IBM Peer in preference to samba for my local network so it would not work.

As far as lubuntu goes: it is a ubuntu variant and like all "nixes" is a PITA. The final straw is that lubuntu decided to boot to a totally mangled display for no obvious reason and after several reboots with the same result is no longer a VBox guest.

Your comment "I've found Windows 7 setups for networking to be wonky, as it wants to do everything for you, its way, and assumes you have the IQ of a carrot." seems to apply to most nixes as well.


Regards

Pete


Heh Pete!

I use IBM File and Print Client as well for Peer, in addition to Samba (after I found out you could actually install it still as one of the ArcaOS options the second or third time around as I needed Virtual PC). I haven't tried setting up either, for my PALNET network, to attempt to read from a VBOX VM, but should see if I can set aside some time to do that,  at some point., as a test.

I suspect it would actually require the Virtual Switch that Valery referred to (TAP.OS2) at least for PEER.

The last time I dappled in "nixes" was setting up XFree86 (various versions) about a decade (or was it more?) ago, which was an arcane exercise in itself. But at least it introduced me to the wonderful world of Unix file hierarchies and frameworks, so installing numerous DLL's to various folders manually, prior to YUM/RPM and ANPM didn't seem so strange. I found the display drivers there under XFree86 to be a pain as well, but it worked.  There was a project "EVERBLUE" after that, but I don't think that went anywhere.

I nixed the "nixes" emulation environment at the time, as I didn't find anything I couldn't do natively OS/2 or eCS wise, or through VirtualPC. As mentioned, I'm not a programmer, just an expert user, and a businessman.

And regarding Windows 7, my last experience was trying to get a ThinkPad T510 with Windows 7 to be recognized on my network by my OS/2 derived machines, as it constantly wanted to steer me towards "Windows Homegroup". After much reading (and a few curses in various exotic languages), I was able to do a lot of manual network set-up. so that everything is seen and accessible now, through NetDrive - but it wasn't easy.

If there's anything to be said for staying in this community after all these years, its that its made us all very versatile, creative ........... and most of all adaptable.

Best!

M
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on May 14, 2018, 05:57:28 pm
Vbox appears to require

SET LIBCX_HIGHMEM=2

In order to start. If  SET LIBCX_HIGHMEM=3, or the line is REM out in config.sys, then I get the error below

-----------------------------------------
Failed to create the VirtualBoxClient COM
The application will now terminate
Callee RC: NS_ERROR_ABORT (0x80004004)
-----------------------------

I am using libc 0.6.6-36, libc-devel 0.6.6.36
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on May 14, 2018, 06:15:54 pm
Good Morning Doug!

Vbox appears to require

SET LIBCX_HIGHMEM=2

In order to start. If  SET LIBCX_HIGHMEM=3, or the line is REM out in config.sys, then I get the error below

-----------------------------------------
Failed to create the VirtualBoxClient COM
The application will now terminate
Callee RC: NS_ERROR_ABORT (0x80004004)
-----------------------------

I am using libc 0.6.6-36, libc-devel 0.6.6.36

Actually the SET LIBCX HIGHMEM statement is required for LIBCX, not LIBC, Doug, and only for versions 0.6.1-1 and above (I don't know if that version is still experimental or not now - originally I got it by installing the netlabs-exp repository access from ANPM, which is available through ACCESSORIES on ANPM - don't know if your just using RPM-YUM or ANPM (free from Arca Noae).

The previous version of LIBCX is 0.6.0-1 and didn't require that statement.

Also, I'm not sure what SET LIBCX HIGHMEM=3 would do, as SET LIBCX_HIGHMEM=2 is the correct statement, so you might want to examine and change that as well.

Best !

Mark
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on May 14, 2018, 06:30:43 pm
BTW the most current version of LIBCX is 0.6.2-1 which does work with the SET LIBCX HIGHMEM=2 statement.

Even with that and the other VirtualAddress Limit, Files and other statements suggested for inclusion in CONFIG.SYS, I still have to use the 15 second delay to open VirtualBox. But after nearly three weeks of constant use, it still is remarkably stable, The only thing I've found that is an on-going issue is the cursor in any of the VM windows being located about 1/16 inch below what I think I'm pointing at (ie. the cursor symbol seen actually when clicked can activate something below where I'm wanting to click).

This seems to be an offset problem and not a delay problem, so it may be present in all versions of VBOX (possible a Guest Additions problem?).

Otherwise I don't really have much more to say generally - I did finally get and install Windows 7, and will provide some observations with that later in the week.

Best!

Mark

Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on May 14, 2018, 08:52:44 pm
Mark,

I had the previous version of LIBCX. Vbox started with that version as long as, and only when,
SET LIBCX_HIGHMEM=2 was in my config.sys.

I have been having an OS2KRNL trap E every morning when I boot up my machine, and sometimes during the day when rebooting: ticket 1888 on ArcaNoae tracker.  My current method around the trap is to reset to VGA (blob -> F3), boot up,  and then reinstall SNAP. Then I am good most of the time until the next morning.  I have tried a number of other methods of booting past the trap: changing VIRTUALADDRESSLIMIT, rearranging the order of the LIBPATH, commenting out various devices, etc. But the reset to VGA is the only thing that has worked consistently.

This morning I tried REMing the line SET LIBCX_HIGHMEM=2, and it turns out that allowed me to boot past the trap. I am going to leave it REMed for a day or two to see if that stops the trap, but that means I can't use vBox for that period of time.

The reason I reported the "issue" is there seemed to be some debate on whether
SET LIBCX_HIGHMEM=2 was actually required or not to run vBox. On my machine vBox will NOT run unless SET LIBCX_HIGHMEM=2.

As for the =3 setting, I found this on Github
---------------------------------------
It can also be controlled by the LIBCX_HIGHMEM environment variable as follows:

    LIBCX_HIGHMEM=0 forces the low memory area to be used for the regular C heap (for testing purposes).

    LIBCX_HIGHMEM=1 forces the high memory area to be used for the regular C heap (the default mode, the same as when this variable is absent).

    LIBCX_HIGHMEM=2 restores the original kLIBC voting process and disables any LIBCx intervention (for compatibility with old applications).

    LIBCX_HIGHMEM=3 is the same as LIBCX_HIGHMEM=1 but a check is made using the kLIBC voting process if there is any DLL that votes against high memory usage (i.e. it's built without -Zhigh-mem). If there is such a DLL, or if the EXE itself is built without -Zhigh-mem, then LIBC will terminate the process using abort() before entering main() with a respective message written to the standard output. This mode is intended for safety checking purposes.
------------
Title: Re: VirtualBox v5.0.51 Testing
Post by: Mark Szkolnicki on May 14, 2018, 10:40:05 pm
Hi Doug!

Mark,

I had the previous version of LIBCX. Vbox started with that version as long as, and only when,
SET LIBCX_HIGHMEM=2 was in my config.sys.

I have been having an OS2KRNL trap E every morning when I boot up my machine, and sometimes during the day when rebooting: ticket 1888 on ArcaNoae tracker.  My current method around the trap is to reset to VGA (blob -> F3), boot up,  and then reinstall SNAP. Then I am good most of the time until the next morning.  I have tried a number of other methods of booting past the trap: changing VIRTUALADDRESSLIMIT, rearranging the order of the LIBPATH, commenting out various devices, etc. But the reset to VGA is the only thing that has worked consistently.

This morning I tried REMing the line SET LIBCX_HIGHMEM=2, and it turns out that allowed me to boot past the trap. I am going to leave it REMed for a day or two to see if that stops the trap, but that means I can't use vBox for that period of time.

The reason I reported the "issue" is there seemed to be some debate on whether
SET LIBCX_HIGHMEM=2 was actually required or not to run vBox. On my machine vBox will NOT run unless SET LIBCX_HIGHMEM=2.

-----------------------------------------
Failed to create the VirtualBoxClient COM
The application will now terminate
Callee RC: NS_ERROR_ABORT (0x80004004)
-----------------------------


Which video card are you using with SNAP, out of curiousity? Sounds like cold reboots TRAP but an initial reset, reinstall and warm reboot works. Perhaps memory clash issues?

As you know I've been using this operating system in its various incarnations for years, but am not technically versed enough in all the memory, programming or kernel intricacies. However it does sound, as a layman, like LIBCX through SET LIBCX HIGHMEM=2 and probably SNAP are fighting for the same high memory or shared memory addresses at boot-up.

I'm using an ASUS licenced ATI 7850 card in this machine, but the version of SNAP provided by ArcaOS trapped the machine when I booted using this card. I use Panorama, latest version (v1.09), as substitute, as all I really do with the card on ArcaOS / eCS / OS/2 is use apps that really have no high performance video requirements (I do have a Windows gaming partition also on this machine, which really uses the high resolution and GPU speed capacities of the card with card specific drivers installed.

Panorama works well enough, but there are times when I have some wonky issues with different programs and Panorama (artifacts / video corruption), most notably Seamonkey, but good enough for what I need.

The error message you get as per the end of your quote above (Failed to create the VirtualBoxClient COM) sounded familiar so I looked back on the VirtualBox Problem thread. if you read from Petes reply #13 on, that problem was solved using the updated LIBCX (at that time v0.6.1-1) and SET LIBCX HIGHMEM=2 used together.

You might want to try upgrading to the newer LIBCX (0.6.2-1) and try the SET LIBCX HIGHMEM=2 or 3 again, to see what happens, or whether anything related to the TRAP is solved. Otherwise I can't think of any other bright ideas myself at the moment, other than potentially using Panorama with your card instead of SNAP, if your willing to live with that (and it doesn't cause other problems with the OS).

Best!

M
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on May 15, 2018, 02:23:49 am
I am using a ATI Radeon PCIe x600 card. Fantastic card, very cheap, quick.

I thought it was related to SNAP because the trap happens after the SNAP splash screen and before the screens turn blue and then the desktop appears. Which is what lead me to revert to VGA and reinstall SNAP.

ArcaNoae thinks it is a timing issue and not a SNAP issue, and my reverting to VGA is effecting the timing of something on startup. There are a number of other people who have gotten the same Trap at the same location.

The trap happens every morning when I first boot up, but sometimes also happens on reboot during the day. I will keep the SET LIBCX_HIGHMEM=2  REMed out for a day or two if I get no traps so I can report back to ArcaNoae.

I now don't THINK there is a conflict between SNAP and Vbox that cause the trap. But there are some memory considerations when using SNAP and vbox, which were talked about on this thread or the other VirtualBox thread. Basically: SNAP will consume lots of memory it apparently doesn't need, which can affect vbox starting and running. The solution is to limit the memory consumed by SNAP with the GAOPTION VIDMEM xx command. I am using 32 for the xx, which limits SNAP to 32MBs. This works ok with my SNAP dual monitor setup when using 64k colors.

If you decide to try the Radeon x600 PCIe there are two considerations to keep in mind. The card is a half height card and you probably want to make sure it comes with full height bracket. Otherwise you have to replace the bracket - which is easy if you have the full height bracket. The other consideration is the card has a weird connector on the back, a DMS59 connector, and you need the cable set to go with it: male DMS59 to 2 female VGA connectors, or male DMS59 to 2 female DVI, or male DMS59 to 1 VGA and 1 DVI. It appears any card will work with any cable set. The part number seems to indicate the bracket height, the memory on the card, and the cable set included with the card, but you can switch the cable sets around and they still work.

The card runs between $5 to $10 on ebay. If it doesn't include the cable you can get the cable for about $5 to $10 on Amazon or ebay.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Dariusz Piatkowski on May 21, 2018, 04:04:11 pm
Hey Doug,

...The trap happens every morning when I first boot up, but sometimes also happens on reboot during the day...

Sorry if I missed it in one of the previous thread posts, but what module are you trapping in?

I'm asking because I had something similar happen to me when experimenting with the SNAP and Panorama drivers. I was switching between them (somewhat of a re-install, but not a clean 'by-the-book' one) and ended up trapping right during a re-boot. Took me a while to research but I ultimately tracked this one down to a config file (the name of course escapes me now, but I probably have it somewhere in my notes).

Soo...I'm hoping that your module name may ring a bell here?
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on May 22, 2018, 06:15:34 am
Dariusz,

I was trapping in OS2KRNL.

I submitted a bug report and got a response and a fix from ArcaNoae on Sat.

The short version is

"The trap is occurring when PM is attempting to load LEDPANEL.FON which is
installed by PMDCALC.WPI.  The font file itself is OK, but it exposes a defect
in _ldrGetMte..."

So far the fix has been working.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Andi B. on August 05, 2018, 12:12:16 pm
> It seems "enable VT-x / AMD-V" are enabled by default in the settings. I was not sure about that previously in my posts to Valery in the thread "VirtualBox Problems" as the original VM I had created (Windows XP) was done under Windows 7 and Windows VirtualBox v5.0.22, so I wasn't sure if I had fiddled with the settings at some point.

No, I just created the new VM from scratch, VT-x/AMD-V is disabled. In any case, disable it if it enabled, for some weird reason (unless you have one of the latest OS/4 kernels).
I came over this again as I set up a new VM with a FreeBSD image. VT-x/AMD-V was enabled here. This is with v5.0.6_OSEr143. A slightly newer version as the official linked at http://trac.netlabs.org/vbox/wiki.

I know this thread is about v5.0.51 but I think VBOX for OS/2 lacks some update on the official wiki page (and the included readme). Although I'm running some VMs inside VBOX I didn't remember the proper settings anymore. Valery please update the wiki page. Especially the necessary/recommended settings for VT-x, IO-APIC, PAE/NX, preferred network mode (which one works reliable), best network adapter setting (I read somewhere some Intel server card should be choosen), ... Of course settings for ArcaOS/eCS are needed. Special instructions for unofficial kernels may need an own paragraph. It's really hard to read all postings to find them out. At least a link at http://trac.netlabs.org/vbox/wiki to the current (updated) readme would be much appreciated. Thanks.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on August 05, 2018, 04:06:42 pm
2Andi B.:

v.5.0.51 is an unofficial leaked version (though, it works quite well). I uploaded it for someone for testing purposes. So, it is not listed on wiki page and it does not include README. README is a bit updated, but not significantly, so a README from latest official version can be used. The current README is always located in the SVN repository in sources: http://trac.netlabs.org/vbox/browser/trunk/doc/ReadMe-OS2.txt (note that SVN trunk is a bit newer version, not yet working and not yet released).

VT-x/AMD-V can be enabled in newly created VM if your system supports hardware virtualization. Though, if it is not the OS/4 kernel, hardware virtualization will not work and the VM will likely panic, if you enabled it. So, if you don't have the OS/4 kernel or your system does not sipport VT-x/AMD-V (like my main system, I have Athlon 64, single core Socket 939 without AMD-V), you should disable it too. You simply need to uncheck the checkbox AMD-V/VT-x. PAE/NX can be checked, it should always work, and doesn't caause VM to crash. IO-APIC is recommended to be disabled for WinXP guest. Otherwise, it may be useful in some cases (depends on a guest OS). NIC depends on guest OS too (for OS/2 guest this may be Intel 1000/Pro MT, as well as AMD PCNet). The only network sharing mode working is simple NAT. TAP driver is currently broken. OS/4 kernel is needed for hardware virtualization support. Note that the official version avaiable for download from the wiki page, does not include hardware virtualization. Unofficial v.5.0.51 supports it. Nothing special is needed. Just enable AMD-V/VT-x in BIOS and in VBox settings. Otherwise, these two versions should be the same.

OK, I'll try to update the wiki. I thought it is evident thing that hardware virtualization is unsupported on IBM's kernels. As it never worked on them and it lacks the "CPU rendez-vous" feature which alows to run code on all CPU cores in parallel. This is essential for enabling AMD-V/VT-x. So, it is experimental feature with OS/4 kernels. Of course, VBox enables it if it finds that it is enabled in BIOS and it is supported by your prosessor. But it cannot check which kernel you have installed.

Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on September 26, 2018, 03:35:05 am
The Import Appliance feature never reaches completion but appears to work OK.

I used File - > Export Applicance to export an WinXP virtual machine I had created on an AMD machine.

I used File -> Import Appliance to import that appliance/virtual machine (the WinXP VM) on an Intel based laptop (Lenovo T530i). The import got to 99% 2/3 and never completed. Or I should say the dialog box never got past 99%.

I used WatchCat (a process monitor) to terminate the vbox session.
 
When I restarted vbox the imported "appliance" was there - i.e. the WinXP VM. It started and appears to be running fine. I had to re-import the guest additions, but other than that it is running fine.

So it appears the import works correctly, even though it never gets past 99% on the progress dialog box.

Kudos to all who are working on this project. Very very nice.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on September 26, 2018, 01:56:13 pm
2Doug Clark: I tried to import one appliance, and it worked for me. Probably, it waited for something at the end (some resource was not free). On my machine,I did not observed any wait, it terminated ok. Why do you need to use WatchCat? Is "Close" button not working?
Title: Re: VirtualBox v5.0.51 Testing
Post by: Lars on September 27, 2018, 09:29:32 am
OK, I'll try to update the wiki. I thought it is evident thing that hardware virtualization is unsupported on IBM's kernels. As it never worked on them and it lacks the "CPU rendez-vous" feature which alows to run code on all CPU cores in parallel. This is essential for enabling AMD-V/VT-x. So, it is experimental feature with OS/4 kernels. Of course, VBox enables it if it finds that it is enabled in BIOS and it is supported by your prosessor. But it cannot check which kernel you have installed.

Hi Valery: I have written a device driver that would do "CPU rendez-vous" to set up "SYSENTER" entry point on all cores, see:
http://hobbes.nmsu.edu/download/pub/os2/system/drivers/misc/syscall.wpi
It comes with full source. It basically issues an IPI to all cores and the cores will then run an interrupt routine, in this case, to set up the necessary MSRs on all cores to being able to use the "SYSENTER"/"SYSEXIT" instructions. This driver works with the standard OS/2 kernel.

I think that you could do about the same to set up virtualization on all cores.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on September 27, 2018, 10:25:55 pm
2Lars: I tested your driver some time ago. The problem was that the OS/2 kernel cannot guarantee the required number of GDT selectors. So, it failed to allocate these selectors and I could not test the driver further. So, if you resolved these problems, I could try. Though, I'm working on other projects now, than VBox. And VBox is currently broken, so before I fix it, I cannot work on CPU rendez-vous with IBM kernel (I updated VBox sources to a newer version and something got broken).
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on October 06, 2018, 08:38:50 am
Valery,

I did it twice - the first time I used the close button. For some reason, which I cannot remember now, I didn't think the import worked and so tried a second time, which is when I used watchcat. Looking back it think the import probably worked the first time, even though the dialog never reached completion.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on October 06, 2018, 09:01:30 am
I am using vBox on my Lenovo laptop to run Windows 3.1 programs, since WinOS2 wont run on this laptop.

I installed Improv in a Windows XP virtual machine in vBox. The install hung at the very end when it was creating icons in a Program Group. But it appears that Improv runs OK, except for accessing shared folders.

The Improv File Open dialog does not show any shared folder "drives". And I cannot successfully map a local drive to drive on a remote machine using the native Windows XP method, either to a SAMBA server on the remote machine or throught  NetBEUI (after installing the NETBEUI driver in the WinXP box)
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on October 06, 2018, 09:46:30 am
2Doug Clark:

Regarding the import/export the appliance: On my side, all is ok. Please try providing more details, so I could reproduce the problem on my machine, otherwise I cannot fix it. Are you sure it is OS/2 port problem? Maybe, the same problem is with windows/linux versions. This functionality seems to be generic for all platforms. If so, you need to reproduce it on other platform, supported by Oracle, and complain to them.

Regarding shared folders: Is Improv a win16 program? Are you sure WinXP provides support for network drives for Win16?

Are you sure network works in VBox at all? Is network enabled in VBox? By default, it uses NAT networking. I'm not sure if VBox supports NetBEUI. For Samba case, it may work only in case you have network settings the same as in host network: the same subnet mask, IP addresses from the same subnet, the same workgroup. Can you reach the machines of your host network from the guest Win32 programs? (not Win16 programs). Is Internet accessible from the guest system?

PS: I suspect that with NAT networking, you'll not be able to access the SMB network drives from the host network. You'll need a fully functional network support in VBox. There's TAP driver support in VBox, but currently it is not working. The only working option is NAT networking.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on October 06, 2018, 09:40:32 pm
Valery:

I imported the same "appliance" (virtual machine) in Windows 7 vBox and it imported OK.

I tried to import the same appliance on an AMD desktop running AOS (as opposed to the Intel Lenovo laptop) and it "hung" at what appears to be the same point - the very end. (First attachment.)

Clicking on the red X - to stop the process, doesn't do anything. (Second attachment) The status/dialog changes to "canceling" and then never goes away.

However after killing the VirtualBox process, and then restarting vBox, the imported appliance/virtual machine is there, and runs correctly.

Since it appears to work I am ok. If you are sitting on a beach soaking up some rays and get bored listening to the waves lapping on the sand and want to work on something then perhaps it is worth fixing. Otherwise maybe it is more appropriate to note in a read.me or something.

Improv is a Windows 16 bit application. I installed it a few minutes ago on a laptop running Windows XP and it can see and access network drives. (third attachment). But when running in a Windows XP virtual machine, the shared folder drives do not show up in the File - Open dialog box. That same issue is also true in Improv running in Windows XP in vBox on Windows 7.

I cannot share a network drive inside of a Windows XP virtual machine using either Samba or Netbeui. Now that I have had some sleep I seem to remember you, or someone else, saying that doesn't work. And with shared folders I really don't need to directly share a drive using networking facilities.  I only tried that because the shared folders/drives did not show up in the Improv File Open dialog.

The workaround is to copy the file(s) I need to use with Improv to a local drive in the virtual machine and access them from Improv there. PITA but it works.

(I will point out that FrameMaker v 5.5 running in Windows XP mode (i.e. Microsoft Virtual PC) on Windows 7 occassionally has problems accessing shared folders/drives out.)

If you do want to attack either problem and there is some information I can send you then please let me know.

In the interest of documenting what happens for the one or two other people in the world who might want to run Improv on WinXP on vBox on OS/2 - I will note that the Improv install hangs at the very end - necessitating rebooting the virtual machine. (attachment 4). But once the machine is rebooted Improv appears for work fine - other than the whole shared folder deal. And you probably want to run IMP20W.EXE instead of IMPROV.EXE; IMPROV.EXE has some issues in WinOS2 that IMP20W.EXE doesn't - so I assume the same might be true on Windows XP.

Finally I will repeat that vBox on OS2 is pretty darn impressive. The problems I have run against are pretty minor compared to the function and preformance of the product. Thanks again.



Title: Re: VirtualBox v5.0.51 Testing
Post by: Dave Yeo on October 07, 2018, 01:54:26 am
Another option for running Win3.1 may be DosBox though our port might need updating. Advantages include being able to mount OS/2 directories. I've never tried it but have seen it talked about on the net. Of course you need Win3.1 to install.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Doug Clark on October 09, 2018, 06:35:13 am
Thanks Dave for the suggestion of DosBox.

My observations on DosBox are:

1) Once DosBox starts you can mount a network drive and have that drive accessable inside DosBox, which also means accessible inside a Windows 3.1 application running in Windows 3.11 running in DosBox.

2) While DosBox has a mount command to mount a floppy drive, the Lotus Improv install program doesn't think it is a floppy drive. Which means the installer will not run.

I manually installed by copying the files and editing WIN.INI and the other *.INI files.

3) When you click the mouse in the DosBox, the mouse pointer is "captured", just the same as in vBox and Virtual PC without guest additions. Ctrl-F10 releases the mouse pointer.

4) I found a Windows 3.11 with SoundBlaster installed as a *.rar packaged for DosBox - which worked great and saved a lot of time. I believe all the Warp 4, eComStation and AOS licenses come with license for Windows 3 - so I don't think licensing is an issue when using that method to install Windows 3 that is running in OS/2.

5) What I ended up with is my  Windows application running in Windows, running in Dos. In all its glory, and all its agony. Meaning I saw a divide by zero error in a worksheet that never experienced that before. And at one time Windows wouldn't load the application saying it had run out of application memory. Ah - the good old days.

6) OS/2 really is a better windows than windows. It has been such a long time since I ran a real windows 3 machine I had forgotten how well OS/2 runs Windows 16 bit (and some Windows 32 bit) programs.

7) What you lose in both vBox and DosBox when running windows 3 apps is the ability to cut and paste between windows and OS/2 applications, along with DDE and named pipes. Along with a truly seamless Windows application running on the OS/2 desktop.

But both vBox and DosBox offer the ability to run Windows 3 applications on my lenovo laptop that currently will not run WinOS2.

I don't think you can print - at least directly - from a Windows 3 application running in DosBox. You could create a PDF on a mounted drive and then print that PDF from OS/2.

With vBox running WinXP running Improv I can print from Improv directly to my color HP m452dn printer, which only has drivers for Windows XP and above.

DosBox on OS/2 is pretty impressive, and it will probably run games, and perhaps some applications, that vBox running Windows XP wont. But for Win 3.1 applications that need to print I think that vBox running WinXP is probably a better choice.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Dave Yeo on October 09, 2018, 04:28:52 pm
Interesting, thanks for the report as I'd never actually tried it. Luckily here, WinOS2 continues to work, though it complains about unsupported resolution.
Title: Re: VirtualBox v5.0.51 Testing
Post by: R.M. Klippstein on October 17, 2018, 12:41:43 am
Has anybody been able to install VBox guest additions on 5.0.51 using OS4 kernel with ArcaOS 5.0.3? I've tried a couple of .iso's with disastorus results. Looks like there needs to be some updating of VBoxAdditions.iso for ArcaOS 5.0.3!

Thanks for any info or ideas.   klipp
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on October 17, 2018, 01:13:59 pm
2rklipp: Which VBox additions are you trying to use? Self-compiled from latest VBox? Did you used the shared folders IFS and daemon from VBox OS/2 svn repo? If so, then I should say that there is some problem with OS/4 kernel and these additions. I did not yet fixed that problem. The daemon just does not finish automounting of shared folders and hangs. This is not yet fixed. It works with IBM kernels, though.
Title: Re: VirtualBox v5.0.51 Testing
Post by: R.M. Klippstein on October 18, 2018, 02:28:12 am
O.K. Thanks valerius. I'm for now only set up to use OS4 kernel and ArcaOS 5.0.3. I'll set up on another machine using OS4 kernel and ArcaOS 5.0.2 or 5.0.1 and see what happens  there.
Another question, Will all future releases of VBox be available on you're site and no longer on Netlabs?

klipp
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on October 18, 2018, 02:28:05 pm
2rklipp: Svn repository is still on Netlabs. But I have no write access to Netlabs' ftp, so I put binaries on my own one. But 5.0.51 is just a test build. The official one was on BWW's Dropbox page.

PS: No matter if it is Arca OS 5.0.3 or 5.0.2/5.0.1. The problem is with OS/4 kernel (we'll try to fix that, of course, but for now, use any IBM's kernel. Arca's kernels are just patched IBM kernels)
Title: Re: VirtualBox v5.0.51 Testing
Post by: Andi B. on October 18, 2018, 05:42:04 pm
Quote
I have no write access to Netlabs' ftp,
FTP write access was disabled long ago. If you want to upload something you have to use WebDAV. The netdrive plugin works well.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Andreas Schnellbacher on October 18, 2018, 09:34:40 pm
And WebDAV access is not automatically activated if you have write access to SVN and trac.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on October 19, 2018, 02:26:57 am
2Andi B.: I have access to yum repository via WebDAV, but not to the ftp. Anyway, VBox weights very much and I'm not sure if Netlabs' ftp has enough space to host the VBox builds archive. My ftp has enough space, so I'd better use it.

2Andreas Schnellbacher: yes, probably, I also need to activate access to the ftp via WebDAV. I have write access to yum repo, and the password is the same as Trac/SVN passwords are. But access is enabled separately, I think.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on October 19, 2018, 02:33:55 am
PS: I tried the latest OS/4 kernel today, r5283, and now I see that VBox additions are working! But there's still a problem with ssmdd.sys driver hanging on init with that kernel. But if you comment it out, everything is booting successfully and working well. We'll try fixing the problem with ssmdd.sys as soon as possible.
Title: Re: VirtualBox v5.0.51 Testing
Post by: R.M. Klippstein on October 21, 2018, 08:50:36 pm
valerius, I'm trying to get kernel r5283 but it looks like www.gus.biysk.ru is  not reachable, there used to be another site where you could download  OS4 SVN's  bit I lost that during my last H.D. crash. Any ideas as to where I can get it?

Thanks klipp
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on October 22, 2018, 05:07:56 am
2R.M.Kilppstein: It is gus.biysk.ru, without www. But AFAICS, the site is currently down. You could try opening it later.
Title: Re: VirtualBox v5.0.51 Testing
Post by: Valery Sedletski on October 23, 2018, 01:23:55 pm
BTW, gus.biysk.ru is back. They said, the problem was the dead cooler on that machine. It's old (Socket 370/SocketA), so it was hard to find the same.