• Welcome to OS2World OLD-STATIC-BACKUP Forum.
 

News:

This is an old OS2World backup forum for reference only. IT IS READ ONLY!!!

If you need help with OS/2 - eComStation visit http://www.os2world.com/forum

Main Menu
Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - djcaetano

#166

  Since I had installed DaniS506 1.8.0 and 1.8.1 I had noticed that I am not able to access my CD drive. On my Warp 4.52 machine I reverted back to the latest 1.7.x and the CD drive came back; on the eCS machine, though, I was not able to revert to 1.7.x, since the newest ACPI installed require 1.8.x.
  Anyone has any tips or clues about what is happening? Anyone with the same problems?

  Thanks in advance.
#167
Quote from: Saijin_Naib on 2008.02.07, 01:13:34
Here is loopback performance for me. These numbers are NUTS. What is going on here?

  There is nothing nuts about them. They show the TCP/IP stack performance,
without interference of network driver and/or network lags. It seems OS/2's TCP/IP
stack is pretty faster than Windows stack. :)
  Also, the speed increase in OS/2 TCP/IP stack is measured as expected (it's
bigger for bigger packets). Windows TCP/IP stack has a somewhat inconsistent
speed increase (sending data in 4KB packets is slower than sending data in 2KB
packets... and receiving data in 8KB packets slower than receiving data in 4KB is
a weird result).

   If you are asking yourself "what the hell is the meaning of those numbers?", they
represent the maximum transfer speed your TCP/IP stack with the hardware you are
using.

   As a result, if your transfer is way below the speed of loopback, that means the
bottleneck can be the drivers (Windows and/or OS/2), your NICs or your machines'
connection.
   In that case, maybe simply exchanging the network driver will not give a good
comparison between GenMac and a "native" driver, because it's not that easy
to know what is the bottleneck. Since your connection seems to be an
100Mbps one (based on your tests, which had shown a result of 89Mbps), I
believe the bottleneck is the network cable or NIC.

   ( I am not expert in this area, I just had done some experiments in the past...
if I am talking nonsense, please, someone corrects me! ;) )
#168
Quote from: Saijin_Naib on 2008.02.07, 00:59:20
My results as a client, my roomate as host. The images explain the OS in each
...
this NIC is far better than mine because his speeds seem much higher.

  The test shows there is only a little difference between OS/2 and Windows. If you are using GenMac on eCS, then I believe the performance of GenMac is awesome. If you are not using GenMac, then there is already a base for GenMac comparison.
  The fact that made me think was the fact Windows reception seems indifferent to data size... even when using very small packets (like 1KB), which is somewhat weird to me. Anyway, a measured result is what it is. :)
#169
Quote from: RobertM on 2008.02.04, 21:18:25
To test actual performance of the NIC and driver, one would need an isolated, local area setup with a test machine that had known bandwidth capabilities, and use that to test the other machines from. Besides any weird variances created by SpeakEasy (Java, browser, etc) the Internet in and of itself will skew things further (besides, I doubt you or anyone here has a gigabit pipe to the Internet - making it hard to truly gauge what the card or driver can support).
And of course, the card itself, even with the best of drivers - as well as what else is on the same bus eating bandwidth, can often be a bottleneck. That's why some server class network cards are horrendously expensive (as much as a PC costs in some cases).

  This program is really good, but AFAIR it only works for NetBIOS/NetBIOS over TCP/IP. But it is a good test program:

http://hobbes.nmsu.edu/cgi-bin/h-search?sh=1&button=Search&key=netio&stype=all&sort=type&dir=%2F
 
  And it works on Windows too.
#170
General Discussion / Re: Port of K42?
2008.02.01, 15:26:50
Quote from: chennecke on 2008.02.01, 14:47:28

  • Installation ends with the message that a class could not be registered.
  • Research reveals that the package needs three supplemental packages:

    • The dependency is not mentioned anywhere in the WarpIN installation instructions.
    • The dependency is not a requirement for being able to install the WarpIN package as it should be.
    • The mandatory packages are not available in the same place as eSchemes.
Yes... the same here. I had to dig into the file to discover Cairo 1.something was needed (I had only an older version installed). And since Cairo is distributed as a WarpIn package, it would be nice if eSchemes verified if it is installed or not.
   About not being available in the same place, I think this would be not a problem, since the requirement was clearly expressed by WarpIn. It would be just a matter of searching for it in OS/2 World or hobbes (the usual places), or even NetLabs FTP.
  Anyway, this creates some kind of "dependency hell" a la Linux.

Quote from: chennecke on 2008.02.01, 14:47:28

  • De-installing eSchemes removes vital parts of the system instead of restoring them to the previous state.

    • The system is unusable.
    • The precautions mentioned in the installation instructions do not help in restoring a usable state.

  This was the worst part. :(
  Anyway, at least I was able to recover (well, somewhat).

  I would like to add that in any way I am "poking" the authors. I believe they are doing the best and I sheer for it. I just complain a little exactly because I *want* to use their extensions (which are really nice enhancements!), but I can't and sometimes I simply don't know why.
  Why WPS-Wizzard works ok on several places and not on any of my machines? Why eSchemes is not able to work in my machines? :(

  I believe something must be done to manage WPS extensions. For me, these problems are much more annoying than SIQ problems.

Quote from: chennecke on 2008.02.01, 14:47:28
Regarding Eugene's announcements, I'm not talking about a language barrier problem. I'm talking about a flood of half-baked ideas and showing an extremely short attention span. The activity I am talking about has put off developers and resellers to a point where communication groups he posts in have become unusable because people just switch to ignore mode as soon as he says something. That way nothing really happens and good ideas from both Eugene and other people get lost. Now, that's not helping.

  I understand what you are saying, but I understand Eugene's behavior. There are so many (small or big) things to get done and almost no time to do everything! As a hobbyist developer for an old computer system, I also start a lot of projects, many are just stalled... but some are finished at this time. People used to blame me for not finishing several things I started ... and the solution (in that case) was simple: I just stopped to say what I was working with.

   But in OS/2 scene I believe this is not a good thing, because it increases the chance of two people spending time developing the same feature (in case of closed source) and diminishes the number of working hands (on open source projects). Also, beta-testing is important, and it is needed to release something so people can test it.
   The down side is when a closed source project is revealed as "work in progress" and the author stops its development, then no one will start another project to do the same thing nor the project will be finished, and the users will be forever without that feature.

   Well, those were my 2 cents.
[/list]
#171
Quote from: David Graser on 2008.01.31, 12:50:49
I have spent many hours trying to enhanced this package.  To do this, I had to create many new icons.  Hope everyone likes the changes and additions.

   You rock, man! Superb work! I dream with the day I will be able to use it in my system! :D
#172
General Discussion / Re: eCS 2.04 RC4
2008.01.31, 15:21:17
Quote from: Saijin_Naib on 2008.01.31, 06:01:27
Okay, I set Netbios Name Scope string as rh.rit.edu and Local Netbios Name Scope String as Workgroup. ALl other values are set as described, TCP/IP is bound to 0, as is Netbios over TCP/IP. Which should be 1, which should be 0?

  I always place NetBIOS over TCP as 0, and then assign it to net1 on IBMLAN.INI, so NetBIOS over TCP/IP is always 'tried" before plain NetBIOS.
#173
  Hi Cris,

Quote from: Cristiano Guadagnino on 2008.01.31, 14:24:55
Christian, do you have experience with the recent versions of their packages? eSchemes v1.46 is a great improvement over the previous versions, and a HUGE improvement with respect to the tools we previously had to modify eCS's look. We badly need this, even if it is only "looks".

  I had tried to install eSchemes v1.46 in this eCS 2.0RC1 and it caused a real mess: all frame window buttons (close, maximize, etc) vanished (they were working, but vanished) and every dialog control (buttons, radio boxes, check boxes...) stopped to work. The system became unusable and I was not able even of uninstalling it without hack: I had to rename estlrl11.dll, reboot machine, and uninstall eSchemes v1.46. Problem is: it deleted all DLLs (even those from eStyler that come with eCS) and now the eStyler cannot run anymore (it complains about missing dlls). Also, the features provided by eStyler (like gradient colors on titlebars) are not working anymore. And I was not even able to see what eSchemes really does! :(

  The main problem with most WPS enhancers is the fact they do not have a good "uninstall" feature; a good uninstall feature is that once the add-on is uninstalled, the system goes back to the previous state. In the case of eSchemes, the readme says one should backup INIs (I had done that), but backing up INIs cannot help with restoring DLLs. Also, I never suspected eSchemes would replace eStyler. Maybe it is said somewhere (?), but since the program has a different name (eSchemes / eStyler), it was very annoying to verify that I was not able to install a new add-on and, as a result of trying, cannot use the older add-on that was working without problems.

   Just as a statement about WPS add-on problems, WPS-Wizzard also game me some headaches. I tried to install it and I thought it worked... but the system became unstable (beeps, WPS hangs, vanishing icons...), even disabling all features in the WPS Wizzard configuration. When installing a icon scheme, even at 32 bit color mode (using Panorama Video Drivers) the icons were washed-out when displayed in small sizes (in eCenter). I like a lot the features of WPS-Wizzard; its the most impressive WPS feature add-on since xWorkplace - and for that I would like to congratulate the author, which had created good multimedia classes and also superb Audio Data CD/DVD Creator, but I never was able to keep it installed in my three OS/2 systems for the same reason: WPS instability (beeps, hangs etc). I thought it was because I had too much add-ons in my Warp 4.52 CP2's WPS, but this eCS is "vanilla", the only WPS add-ons installed are those that come with eCS 2.0RC1.
   Anyway, the uninstall was not without problems also. The system hanged while restarting the WPS and I had to uninstall it the ugly way: renaming its DLLs, rebooting, running checkini, cleaning all references to its classes and THEN uninstall it.

   I would like to notice that I am *not* complaining about the authors.  ;) They are doing a supreme effort to keep OS/2 / eCS alive, with new functionality and updated look and feel  :) ; but WPS is really tricky: it allows these updates, but there are a lot of checks that must be done before the install program could extend it. Also, the install should save files needed for a correct uninstall process *and* the uninstall should work even when things had gone wrong (since most people uninstall a program like those exactly because something has gone wrong!).

   Maybe serenity systems should spend some time extending WarpIn so WPS add-ons could be installed and uninstalled without those problems. I know how to cope with this mess, if needed I am able to open a class list, delete the offending class, locate and rename faulty DLL files and somewhat restore the system manually, but many users can't and then we are obligated to listen that old story that "OS/2 needs to be reinstalled once every month, which is annoying" ... :(
#174
General Discussion / Re: eCS 2.04 RC4
2008.01.31, 04:59:19
Quote from: Saijin_Naib on 2008.01.31, 03:53:02
Thanks for all the info and help! I will try these shortly, but just a quick comment from my end. Its rather strange, my roomate can access my eCS shares perfectly, without issue, from his Vista and XP boxes, with no reg changes, and no netbios enabled! In fact, every other person on the workgroup I asked could see me and access me, and they did not do any tinkering to get it to work. Weird stuff.

  Very weird.

Quote from: Saijin_Naib on 2008.01.31, 03:53:02
Two questions:
Enter NetBIOS Name Scope String: (TCP/IP domain name? Where would I find this?)

  If your computer is inserted into a TCP/IP network with domain names, then your
computer should have a TCP/IP domain name. For example, my computer is
connect through my provider which is named Virtua. My computer hostname
is SAURON. The TCP/IP domain is virtua.com.br and therefore my computer
can be accessed inside this network using the address sauron.virtua.com.br.
In Windows 98 there was a tab called DNS Configuration on "TCP/IP" properties
with one field called "HOST" (where SAURON would be show, in my computer)
and another field called "DOMAIN" (where virtua.com.br would be shown, in
my computer)

Quote from: Saijin_Naib on 2008.01.31, 03:53:02
Enter Local NetBIOS Name Scope String: (Netbios/p2p domain name? What is this?)

  Usually it is the name of your Workgroup or something like that on Windows.

Quote from: Saijin_Naib on 2008.01.31, 03:53:02
Also, does it matter that my IBM OS/2 NetBIOS Over TCP/IP and IBM OS/2 TCP/IP are both set to number 0?
thanks!

  I had bad experiences on that. Usually you should have NetBIOS over TCP/IP as 0 and NetBIOS as 1. Notice, however, that MPTS Configuration do not update things correctly when changing protocol order. Make a backup of your /IBMCOM/PROTOCOL.INI file befora changing things around. After placing NetBIOS over TCP/IP as 0 and NetBIOS as 1, the following section shoud be present on your PROTOCOL.INI:

[NETBIOS]
   DriverName = netbios$
   ADAPTER1 = tcpbeui$,0
   ADAPTER2 = netbeui$,1

  Notice, however, that MPTS *never* updates the LAN Requester/FPClient. So, look at the file /IBMLAN/IBMLAN.INI and verify that the following is defined:

[networks]
  net1 = TCPBEUI$,0,LM10,34,100,14
  net2 = NETBEUI$,1,LM10,34,100,14

  The important numbers are that following TCPBEUI$ (0, in this case) and NETBEUI$ (1, in this case). They should match the values of PROTOCOL.INI. Also, verify if the following is defined in this IBMLAN.INI:

  wrknets = net1,net2
  Computername = [your computer name]
  Domain = [netbios domain / workgroup]

And...

  srvnets = net1,net2
  SRVCOMMENT = [your computer name]

And also verify if the tag:

  srvhidden = no

  Is set to NO. The default is YES.

#175
Hardware / Re: Freezing of the WPS
2008.01.31, 03:34:50
Quote from: RobertM on 2008.01.31, 03:24:03
Yup... removed all registry keys, the ft2lib.dll and libc06b4.dll, reinstalled, hang b4 desktop loads...

  Correct it the ugly way: boot to command line and rename ft2lib.dll. I had to do this before, but not with Innotek font engine (I am using 2.60 for a lot of time and never had any problems with it - Warp 4.52 refresh 2)... the problem was when I tried to install FreeType/2.

Quote from: RobertM on 2008.01.31, 03:24:03
It seems I may have had 2.40 running before (and archived copies of the newer DLL)... which would indicate I tried this before with no success. Anyway... I reinstalled 2.40 and that is up and running fine... so I am at a loss. I might try re-installing 2.60b and then moving the libc dll someplace that is first in the path list - maybe it is conflicting with all the other versions of libc6 installed for other stuff. It is the only copy of libc06b4.dll installed... but maybe it is conflicting with one of these?
libc06.dll
libc061.dll
libc062.dll
libc0621.dll
libc063.dll
libc0631.dll
libc0632.dll
libc0633.dll
libc06b2.dll
libc06r1.dll
Specifically maybe r1 or b2? or 6.1+? Got me... but for now I have font support again, so I will mess with it some more after I get some work done...

  Difficult to say. POPUPLOG.OS2 hasn't any clues?
#176
General Discussion / Re: eCS 2.04 RC4
2008.01.31, 03:29:13
Quote from: Saijin_Naib on 2008.01.31, 02:50:21
Alright, I will set myself to H-node again and put the WINS server value back in. Can you give me an overview of what my values SHOULD be in that page? I will post a screenshot shortly.
Edit: Screenshot Time

  Well, let me see... I will send you the values I use (I can talk with another WinXP computer in my home network, but it's a point-to-point network, no server):

Node Type: H-Node
Enable SLIP/PP Interface:
NetBIOS Name Server address: [your WINS server]
Backup NetBIOS Name Server address: [your WINS server]
NetBIOS Datagram Distributor address: [your WINS server]
Backup NetBIOS Datagram Distributor address: [your WINS server]
NETBIOS trace level: 0
Maximum Sessions: 225
Maximum Commands: 219
Maximum names: 44
GDT selectors: 15
Full buffer datagrams: NO
Query timeout: 500
NETBIOS retries: 2
Number of names in remote name directory: 1000
Minutes between remote name directory purgues: 0
Preload remote name directory: YES
Maximum number of name-ip address pairs in names file: 100
UDP-frame descriptors: 20
TCP-frame descriptors: 100
Enter NetBIOS Domain Scope String: [your TCP/IP domain name... p.e: aol.com]
Enablement of DNS lookup of RFC-unencoded name: 2
Enter Local NetBIOS Name Scope String: [your NetBIOS/P2P domain name]
Seconds to pause between TCP/IP interface checks: 300

  The most important differences are the domain names that must be added, and the Enablement of DNS lookup equal to 2 (this instructs the protocol to try to resolve names by NetBIOS *and* by TCP/IP DNS). The Preload remote name directory to YES and Maximum number of name-ip address pairs in names file are for using with /IBMCOM/RFCNAMES.LST, something that is very usefull if some machines have always the same IPs... this avoid broadcasting to determine machines and thus reduces the NetBIOS response time.
   There is also /IBMCOM/RFCBCST.LST, where you can define the IPs you want to be broadcasted, limiting the broadcast scope and making things faster.

  In my case, I have a wireless router which acts like a firewall server. My RFCNAMES.LST has only one entry, which has my router IP and its name. This router assigns dynamic IPs for machines in my local network, but this IPs re limited from 192.168.1.100 to 192.168.1.115, whose are listed in /IBMCOM/RFCBCST.LST.

  I don't know if this will solve your problems. It seems to me that some kind of black magic is necessary to make LAN Requester and File and Print Client to talk to Windows machines and domains. I had tried in different configurations with mixed results. In some places I could see everything. In others only could add printers and folders "forcing" them (asking to WPS to create icons). And has been a long time since a Windows could access my computer (the last one was running Win98). I can almost always access WinXP, but WinXP never can do anything with OS/2 shared folders. They are displayed, but when one tries to access it, a password is asked and I never was able to discover it.
  Also, I discovered that WinXP refuses to work with OS/2 Warp Server. Win98 can talk to OS/2 Warp Server (but it needs a lot of tampering with lmhosts file on Win98), but not any Windows NT since Win2k (including XP and, AFAIK, Vista).

  Ah, remember to enable NetBIOS protocol in your WinXP machines (I don't know why, this seems to make things easier) and also to change every appearance of "lmannounce" in your Windows Registry from 0 to 1.
#177
Article Discussions / ACPI deep testing
2008.01.31, 03:02:27
http://www.os2world.com/content/view/16417/2/
QuotePosted by Martin Iturbide - Friday, 11 January 2008



  I was running SMP-safe UCONV in my W4 kernel machine (uniprocessor) and everything was going fine... until I discovered ScummVM not being able to run anymore, crashing on UCONV.DLL:

01-30-2008  22:55:23  SYS3175  PID 01fb  TID 0001  Slot 00dd
J:\SCUMMVM\SCUMMVM\SCUMMVM.EXE
c0000005
1ff200c9
P1=00000001  P2=00000000  P3=XXXXXXXX  P4=XXXXXXXX 
EAX=0286fc24  EBX=0286fe78  ECX=0286fe86  EDX=0286fc17
ESI=00000000  EDI=00000000 
DS=0053  DSACC=f0f3  DSLIM=ffffffff 
ES=0053  ESACC=f0f3  ESLIM=ffffffff 
FS=150b  FSACC=00f3  FSLIM=00000030
GS=0000  GSACC=****  GSLIM=********
CS:EIP=005b:1ff200c9  CSACC=f0df  CSLIM=ffffffff
SS:ESP=0053:0286fc2c  SSACC=f0f3  SSLIM=ffffffff
EBP=0286feb4  FLG=00012217

UCONV.DLL 0001:000000c9

   Anyone had better results? Are this new UCONV supposed to run only on multiprocessor machines with SMP kernel?
#178
http://www.os2world.com/content/view/16714/2/
QuotePosted by Martin Iturbide - Tuesday, 29 January 2008



The link for new OS/2 version in ScummVM project website is wrong. It points to 0.10.0 version.
The link for the new version is:
http://prdownloads.sourceforge.net/scummvm/scummvm-0.11.0-os2.zip?download
#179
General Discussion / Re: eCS 2.04 RC4
2008.01.31, 02:35:02
Quote from: Saijin_Naib on 2008.01.30, 22:07:06
Okay, so Ive tried all of these values. These are from my school's IT service. Apparently, we no longer have NetBIOS name servers, and have not had them for a long time. Which makes we wonder how the hell I had it working that one time?
WINS Server: 129.21.3.95 (wins.rit.edu)

  AFAIK, if truly up and running, WINS Server is able to act like as a NetBIOS Name Server (NBNS).
  I would use this IP as name server for netbios.
#180
Quote from: kimhav on 2008.01.29, 15:21:31
So let's use our heads hear why would actually IBM benefit from releasing the API's? For example I know of customers that still uses OS2 and actually during this last week I got several e-mails from corporates that where asking for help to find information and solutions to problem they've had. The question would rather be asked why would IBM's customers need the API's. As IBM points to their local service representatives; that's a joke. I mean during the last couple of years IBM Sweden have forwarded OS2 requests to Swedish OS2 User Group! Oh, how much money did we get for that from IBM... nada, zero, null. IBM are actually leeching on the community now.

   Yes, I am aware of that. IBM's answer like someone inside IBM support knows what is OS/2. At IBM Brazil, IBM's support says "OS/2 is dead" since 1997. Local service representatives do nothing regarding OS/2. But it seems IBM-US thinks they really do something... or they are really cynic guys.
   This is the problem: it looks it's not possible to reason with IBM, because IBM support and answers seems to live in some kind of dreamland, where their support system really works and we are just a bunch of bums asking for free beer. I hate the way they despise us with generic non-sense answers.

Quote from: kimhav on 2008.01.29, 15:21:31
Then are there anything in the code that we would like to get; or to turn it around; we've asked and said that we would like to get IBM to open source the kernel, wps, ans som code. With the dialogue with MenSys and Netlabs we've earlier said that we should either point out certain portions or patents that we would like to get information about. This information could either be delivered out to a group of people to handle and my suggestion would be either Netlabs or our self, OS2 World Foundation. So the question again; what specific are we in the need of being able to peak and poke around in? I'm not a developer, so please help me out here!

  I don't know, I had never developed an entire Operating System. But I had done myself a lot of reverse engineering and I would say the base OS API (and kernel?) docs would help a lot. Also, the release SOM 3.0 (fully CORBA compatible) and WPS classes code would help  A LOT. :)
  The real thing for WPS is SOM engine. And I believe there is not any third-party patents involved in this code, since I never read anywhere something like that. Also, I had already read somewhere that once IBM licensed SOM 3.0 to another company... I believe its name was Tandem or something alike.

Quote from: kimhav on 2008.01.29, 15:21:31
What specific rights would that be; lots of posting claims the same; but where is the facts? What about trying to figure out what Microsoft would be totally against getting out in the wild. Where are the links, info, etc?

  It is hard to say, but although many things changed, some of actual OS/2 kernel and much of base API *is* from 1.x version, which was fully developed by Microsoft.
  "Looking in the right places", one can find the sources for a really old version of OS/2 and see that almost all of it is copyrighted by Microsoft.

Quote from: kimhav on 2008.01.29, 15:21:31
True, they want to but they can't since they still have customers depending working systems. Problem is that we don't have any large or customers standing behind this petition; not as corporation but sure as single individuals within an organisations.

   They want these systems do not work anymore (unless the owner pays a really big amount of money) so these users pays them to develop a migration plan. :(
   It seems the only language they talks now is money language.

Quote from: kimhav on 2008.01.29, 15:21:31
Well, guess that we have tried and manage to get our fair share of hits with the petition. But, what we really need now is the facts regarding code needed.

  Well, I really don't know, but I suppose the base API/kernel docs (since the source cannot be released) and SOM/WPS is a very good start. It'll be a lot of work, but would help *a lot* the guys working on osFree and Voyager.

Quote from: kimhav on 2008.01.29, 15:21:31
Sure no problem; there are already quite a lot of bounties where some of them should be joined together into one project.
But, to take a really bad quote from Hollywood; help us to help you...

  Hehehe... My only need is a decent OS to play around. I cannot stand the way Linux/Unix deals with the user. And I cannot stand the bugs, bloats and never-stop upgrading life in Windows. :D