OS2 World Community Forum

OS/2, eCS & ArcaOS - Technical => Networking => Topic started by: Dariusz Piatkowski on October 06, 2017, 06:07:29 am

Title: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on October 06, 2017, 06:07:29 am
So here is a weird one...I have been using phone line DSL for a number of years...finally with two teenagers at home and their seemingly insatiable desire for media (well, we all enjoy it once in a while LOL) I took a plunge into the Fibre options.

Today the local cable provider ran a coax cable from the box at the front of the house (fibre backbone) to the house. He installed the modem, I hooked my WifFi router up, all good right???

Well, to my surprise, the OS/2 machine can only fathom a lousy 800KB/s, which is roughly 8Mb/s, whereas the exact same download on my Win7 box literally flies almost 10x faster giving me an avg 7100KB/s, so nearly 70Mb/s.

What gives???

My NIC is classified by PCI utility as "RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller", Device 8168h, that makes sense. I am using the MMRE drivers, and even though this appears to be a hardware issue I am nearly 100% it is not. I can do a xfer between any of my home devices at much faster rates...so the NIC itself and the OS/2 drivers seem to be very capable of moving FAST.

...and so I'm stumped here!!! Where to go next with this?

I have never had to try to troubleshoot this kind of a problem, I suspect maybe some tcp/ip settings need to be changed?

Needless to say, the move to UltraFibre120 seems to have been much of a fizzle on my OS/2 box...at least the rest of the house devices are seeing the benefits...my problem is I do almost all of my computing/on-line work on my OS/2 box!
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on October 06, 2017, 10:17:07 am
Hi Dariusz,

What are the contents of your x:\mptn\etc\inetcfg.ini file ?
If you do not have one, you can create one with "inetcfg -g all" from the command line.
Then place the line "inetcfg -s all" (minus the quotes) in your x:\mptn\bin\setup.cmd file.

It is likely that your TCP/IP window parameters will need to be adjusted, likely by experimenting, remember the OS/2 stack is ancient :o)

Quote
#Inetcfg:   CURRENT   DEFAULT   MINIMUM   MAXIMUM
tcpswinsize   131072   32768   1024   246723   TCP SendWindow Size
tcprwinsize   131072   32768   1024   246723   TCP RecvWindow Size
udpswinsize   41600   9216   1024   246723   UDP SendWindow Size
udprwinsize   41600   41600   1024   246723   UDP RecvWindow Size
lipcstswinsize   20480   10240   1024   246723   LIPC Stream SendWindow Size
lipcstrwinsize   20480   10240   1024   246723   LIPC Stream RecvWindow Size
lipcdgswinsize   8192   2048   1024   246723   LIPC Dgram SendWindow Size
lipcdgrwinsize   8192   4096   1024   246723   LIPC Dgram RecvWindow Size
perfhttp   1   1   0   1   Fast Path HTTP ON/OFF

Also, what is the value of    "ETHERAND_TYPE" in your x:\ibmcom\protocol.ini file ?
I find I talk a lot quicker to windows PC's via TCPIP with the value set to D but if you change it, it must be changed on all other OS/2 computers otherwise they will nolonger talk to each other.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on October 06, 2017, 05:44:17 pm
H Ian!

What are the contents of your x:\mptn\etc\inetcfg.ini file ?

I did not have this file created, however, here are the values I was using:

#Inetcfg:   CURRENT   DEFAULT   MINIMUM   MAXIMUM

keepalive   7800   7800   0   7800   KeepAlive (sec)
tcpswinsize   32768   32768   1024   246723   TCP SendWindow Size
tcprwinsize   32768   32768   1024   246723   TCP RecvWindow Size
udpswinsize   9216   9216   1024   246723   UDP SendWindow Size
udprwinsize   41600   41600   1024   246723   UDP RecvWindow Size
lingertime   120   120   0   65535   Linger Time (sec)
probecnt   8   8   1   8   Number Of KeepAlive Probes
ipforward   0   0   0   1   IP Forwarding flag ON/OFF
mtudiscover   1   1   0   1   TCP Path MTU Discovery ON/OFF
arpkillc   1200   1200   180   15300   ARP Complete TimeOut (sec)
arpkilli   180   180   60   1200   ARP InComplete TimeOut (sec)
lipcstswinsize   10240   10240   1024   246723   LIPC Stream SendWindow Size
lipcstrwinsize   10240   10240   1024   246723   LIPC Stream RecvWindow Size
lipcdgswinsize   2048   2048   1024   246723   LIPC Dgram SendWindow Size
lipcdgrwinsize   4096   4096   1024   246723   LIPC Dgram RecvWindow Size
synattack   0   0   0   1   SYN Attack Flag ON/OFF
syncookie   0   0   0   1   SYN Cookie Flag ON/OFF
firewall   0   0   0   1   FIREWALL ON/OFF
multidefrt   1   1   0   1   Multiple Default Routes ON/OFF
winscale   1   1   0   1   TCP Window Scale ON/OFF
timestmp   0   0   0   1   TCP TimeStamps ON/OFF
realslow   20   20   1   100   TCP TW Q Slow Timeout ticks
perfhttp   1   1   0   1   Fast Path HTTP ON/OFF
tcpttl      64   64   1   255   TCP  TTL (hops)
udpttl      64   64   1   255   UDP  TTL (hops)
icmpttl      64   64   1   255   ICMP TTL (hops)
fragttl      60   60   1   255   Fragment TTL (sec)
reusetw      1   1   0   1   Reuse Time Wait ON/OFF
msl      30   30   1   16384   TCP MaxSegLifeTime (sec)
cc      0   0   0   1   TTCP function ON/OFF
sack      1   0   0   1   Selective ACK ON/OFF



...Also, what is the value of    "ETHERAND_TYPE" in your x:\ibmcom\protocol.ini file ?
I find I talk a lot quicker to windows PC's via TCPIP with the value set to D but if you change it, it must be changed on all other OS/2 computers otherwise they will nolonger talk to each other.

I do not have this in my INI, I believe that is because I turned off [netbeui_nif], only '0 - IBM TCP/IP' is bound to the nic. It is set up this way since I was simply not using any of this functionality on my LAN as I do not have any other OS/2 machines here and to get to the Winx boxes I rely on NetDrive instead.

Alright, for now I am experimenting with the changes you suggested. Should I be looking to re-define the MTU size as well? Right now it is set to 1452, the router is set to 1500, and various on-line places suggest the use of 1492...

Thanks!
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on October 06, 2017, 06:02:13 pm
So here is a weird one...I have been using phone line DSL for a number of years...finally with two teenagers at home and their seemingly insatiable desire for media (well, we all enjoy it once in a while LOL) I took a plunge into the Fibre options.

Today the local cable provider ran a coax cable from the box at the front of the house (fibre backbone) to the house. He installed the modem, I hooked my WifFi router up, all good right???

Well, to my surprise, the OS/2 machine can only fathom a lousy 800KB/s, which is roughly 8Mb/s, whereas the exact same download on my Win7 box literally flies almost 10x faster giving me an avg 7100KB/s, so nearly 70Mb/s.

What gives???

My NIC is classified by PCI utility as "RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller", Device 8168h, that makes sense. I am using the MMRE drivers, and even though this appears to be a hardware issue I am nearly 100% it is not. I can do a xfer between any of my home devices at much faster rates...so the NIC itself and the OS/2 drivers seem to be very capable of moving FAST.

...and so I'm stumped here!!! Where to go next with this?

I have never had to try to troubleshoot this kind of a problem, I suspect maybe some tcp/ip settings need to be changed?

Needless to say, the move to UltraFibre120 seems to have been much of a fizzle on my OS/2 box...at least the rest of the house devices are seeing the benefits...my problem is I do almost all of my computing/on-line work on my OS/2 box!

Same result for me.
Only 6Mbps while under windows, I have 60Mbps

(note: this is why I searched how to speed up firefox)
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on October 06, 2017, 07:31:37 pm
Hi Dariusz,

Manual MTU value depends on what your fibre is using but you are using mtudiscover for MTU so its fine.

You would normally use 1500 for PPPoA, 1492 for PPPoE but it's really depending on your provider. Some will only accept 1492 for PPPoA and 1452 for PPPoE as your on fibre I would expect you to use 1500 and then let mtudiscover look after it. I have no idea if your provider allows jumbo packets or not, I don't even know if OS/2 supports MTU values above 1500 either but normally smaller is better than larger, up to a point.

If you have created an inetcfg.ini file in your x:\mptn\etc directory.

Quote
If you do not have one, you can create one with "inetcfg -g all" from the command line.
Then place the line "inetcfg -s all" (minus the quotes) in your x:\mptn\bin\setup.cmd file.

then you can start by editing the inetcfg.ini file, simply copy whats below to replace the one you have.

Note you should change your keepalive and sack values to reflect what is below as well.

Quote
#Inetcfg:   CURRENT   DEFAULT   MINIMUM   MAXIMUM

keepalive   60   7800   0   7800   KeepAlive (sec)
tcpswinsize   131072   32768   1024   246723   TCP SendWindow Size
tcprwinsize   131072   32768   1024   246723   TCP RecvWindow Size
udpswinsize   41600   9216   1024   246723   UDP SendWindow Size
udprwinsize   83200   41600   1024   246723   UDP RecvWindow Size
lingertime   60   120   0   65535   Linger Time (sec)
probecnt   8   8   1   8   Number Of KeepAlive Probes
ipforward   0   0   0   1   IP Forwarding flag ON/OFF
mtudiscover   1   1   0   1   TCP Path MTU Discovery ON/OFF
arpkillc   1200   1200   180   15300   ARP Complete TimeOut (sec)
arpkilli   180   180   60   1200   ARP InComplete TimeOut (sec)
lipcstswinsize   40960   10240   1024   246723   LIPC Stream SendWindow Size
lipcstrwinsize   40960   10240   1024   246723   LIPC Stream RecvWindow Size
lipcdgswinsize   8192   2048   1024   246723   LIPC Dgram SendWindow Size
lipcdgrwinsize   16384   4096   1024   246723   LIPC Dgram RecvWindow Size
synattack   1   0   0   1   SYN Attack Flag ON/OFF
syncookie   0   0   0   1   SYN Cookie Flag ON/OFF
firewall   0   0   0   1   FIREWALL ON/OFF
multidefrt   0   1   0   1   Multiple Default Routes ON/OFF
winscale   1   1   0   1   TCP Window Scale ON/OFF
timestmp   0   0   0   1   TCP TimeStamps ON/OFF
realslow   20   20   1   100   TCP TW Q Slow Timeout ticks
perfhttp   1   1   0   1   Fast Path HTTP ON/OFF
tcpttl      64   64   1   255   TCP  TTL (hops)
udpttl      64   64   1   255   UDP  TTL (hops)
icmpttl      64   64   1   255   ICMP TTL (hops)
fragttl      60   60   1   255   Fragment TTL (sec)
reusetw      1   1   0   1   Reuse Time Wait ON/OFF
msl      30   30   1   16384   TCP MaxSegLifeTime (sec)
cc      0   0   0   1   TTCP function ON/OFF
sack      0   0   0   1   Selective ACK ON/OFF

The main values you want to play with are:-
tcpswinsize
tcprwinsize
udpswinsize
udprwinsize

You can probably try setting them to the max values and see how it goes.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on October 07, 2017, 02:28:50 am
I played with inetdcfg and found some other tips   ( for information, I have inkoy firewall installed )
I saw that speed measured through flash under windows or html5 doesn't provide the same result while html5 indicates lower bandwide !

Here is what I use now

Quote
#Inetcfg:   CURRENT   DEFAULT   MINIMUM   MAXIMUM

keepalive   300   7800   0   7800   KeepAlive (sec)
tcpswinsize   131072   32768   1024   246723   TCP SendWindow Size
tcprwinsize   131072   32768   1024   246723   TCP RecvWindow Size
udpswinsize   36864   9216   1024   246723   UDP SendWindow Size
udprwinsize   166400   41600   1024   246723   UDP RecvWindow Size
lingertime   120   120   0   65535   Linger Time (sec)
probecnt   8   8   1   8   Number Of KeepAlive Probes
ipforward   1   0   0   1   IP Forwarding flag ON/OFF
mtudiscover   1   1   0   1   TCP Path MTU Discovery ON/OFF
arpkillc   1200   1200   180   15300   ARP Complete TimeOut (sec)
arpkilli   180   180   60   1200   ARP InComplete TimeOut (sec)
lipcstswinsize   40960   10240   1024   246723   LIPC Stream SendWindow Size
lipcstrwinsize   40960   10240   1024   246723   LIPC Stream RecvWindow Size
lipcdgswinsize   8192   2048   1024   246723   LIPC Dgram SendWindow Size
lipcdgrwinsize   16384   4096   1024   246723   LIPC Dgram RecvWindow Size
synattack   1   0   0   1   SYN Attack Flag ON/OFF
syncookie   0   0   0   1   SYN Cookie Flag ON/OFF
firewall   0   0   0   1   FIREWALL ON/OFF
multidefrt   1   1   0   1   Multiple Default Routes ON/OFF
winscale   1   1   0   1   TCP Window Scale ON/OFF
timestmp   0   0   0   1   TCP TimeStamps ON/OFF
realslow   20   20   1   100   TCP TW Q Slow Timeout ticks
perfhttp   1   1   0   1   Fast Path HTTP ON/OFF
tcpttl      64   64   1   255   TCP  TTL (hops)
udpttl      64   64   1   255   UDP  TTL (hops)
icmpttl      64   64   1   255   ICMP TTL (hops)
fragttl      60   60   1   255   Fragment TTL (sec)
reusetw      1   1   0   1   Reuse Time Wait ON/OFF
msl      30   30   1   16384   TCP MaxSegLifeTime (sec)
cc      0   0   0   1   TTCP function ON/OFF
sack      1   0   0   1   Selective ACK ON/OFF
 

After some new test using link (html5) :
http://speedsmart.net/

I could get 25Mbps ( better than 6Mbps )
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on October 07, 2017, 04:35:48 pm
Hi Ian!

First of all, big Thank You for putting me on the right path here. The results so far are great, let me explain step at a time below...

...
Manual MTU value depends on what your fibre is using but you are using mtudiscover for MTU so its fine.

You would normally use 1500 for PPPoA, 1492 for PPPoE but it's really depending on your provider. Some will only accept 1492 for PPPoA and 1452 for PPPoE as your on fibre I would expect you to use 1500 and then let mtudiscover look after it. I have no idea if your provider allows jumbo packets or not, I don't even know if OS/2 supports MTU values above 1500 either but normally smaller is better than larger, up to a point...

I re-set the value to 1500 in my \mptn\bin\setup.cmd. My old ISP which utilized PPPoE over DSL line specifically requested 1452, the new guys...well, I did not get any info because the tech doing the install was very tight lipped..LOL, he specifically asked me to call their tech line instead.

...
If you have created an inetcfg.ini file in your x:\mptn\etc directory.

Note you should change your keepalive and sack values to reflect what is below as well.
...

Why do you suggest that SACK be off? I am basiing this on my understanding of that flag...the following article does a pretty nice job of explaining it's functionality => http://packetlife.net/blog/2010/jun/17/tcp-selective-acknowledgments-sack/ (http://packetlife.net/blog/2010/jun/17/tcp-selective-acknowledgments-sack/). In short, if both the client and server support this, when an actual packet is dropped this will allow the re-transmission of ONLY the dropped data, as opposed to a full payload from the dropped point onwards until the end of the conversation. If successful this implies less data being re-sent.

Regarding the keepalive, I will change it to sync it up with my primary tcpip apps, that being Firefox and Privoxy. Specifically:

1) network.tcp.keepalive.idle_time 600

Beyond the above, what remains is to play with the sizes of the values for these key parameters you list below:

...
The main values you want to play with are:-
tcpswinsize
tcprwinsize
udpswinsize
udprwinsize
...

So far here are the results for the following settings:
tcpswinsize     131072
tcprwinsize     131072
udpswinsize     41600
udprwinsize     41600

My FF based downloads are now moving at a speedy 50 Mb/s, a HUGE improvement from where I started!!!

Various internet connection speed test report a steady 90-95 Mb/s rates, this is substantially better as well.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on October 07, 2017, 05:54:56 pm
Ian and Remy,

Guys, another update...I thought I would max out the settings with the goal of achieving highest performance...then if I start running into problems I would dial things back a step at a time.

So here are my current applicable INETCFG.INI settings:

#Inetcfg:       CURRENT DEFAULT MINIMUM MAXIMUM

keepalive       300     300     0       7800    KeepAlive (sec)
tcpswinsize     246723  32768   1024    246723  TCP SendWindow Size
tcprwinsize     246723  32768   1024    246723  TCP RecvWindow Size
udpswinsize     246723  41600   1024    246723  UDP SendWindow Size
udprwinsize     246723  41600   1024    246723  UDP RecvWindow Size
lingertime      120     120     0       65535   Linger Time (sec)
ipforward       0       0       0       1       IP Forwarding flag ON/OFF
timestmp        0       0       0       1       TCP TimeStamps ON/OFF
msl             30      30      1       16384   TCP MaxSegLifeTime (sec)
sack            1       0       0       1       Selective ACK ON/OFF

...the xxxwindsize have been set to MAX values.

I still need to research a bit to understand the impact of the following:

lingertime
ipforward
timestamp (I think this may be used by SACK, not sure though)
msl

....and drummroll for the results please!!!  ;D

1) http://speedsmart.net/ (http://speedsmart.net/) => 87 Mb/s
2) http://testmy.net (http://testmy.net) => 40 Mb/s
3) http://beta.speedtest.net (http://beta.speedtest.net) => 91 Mb/s
4) FF download => 5000 KB/s, so roughly sustained 50 Mb/s

All in all, I would say that is basically what I'm seeing on my Winx boxes.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on October 07, 2017, 06:57:34 pm
ipforward is for using your box as a router. I need it here (set in \mptn\bin\setup.cmd) as Injoy on my computer serves the household with NAT
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on October 07, 2017, 07:15:27 pm
Hi Dariusz,

Glad your speed has improved :)

Quote
Why do you suggest that SACK be off? I am basiing this on my understanding of that flag...the following article does a pretty nice job of explaining it's functionality => http://packetlife.net/blog/2010/jun/17/tcp-selective-acknowledgments-sack/. In short, if both the client and server support this, when an actual packet is dropped this will allow the re-transmission of ONLY the dropped data, as opposed to a full payload from the dropped point onwards until the end of the conversation. If successful this implies less data being re-sent.

SACK caused a leak in the TCP/IP stack, I assume it still does as the stack hasn't changed for many years. With it set on I was never able to get my server running longer than 2-4 weeks (from memory), with SACK off by server remains up 24/7 with the only downtime due to power failures. Before moving to were I currently reside the server was up for periods of a year with it being turned off for cleaning and swapping to the backup server which would then take over for the next year. I could/maybe change it and see if it's still relevant but its below by could be bothered threshold at present, maybe Christmas time.

Quote
Regarding the keepalive, I will change it to sync it up with my primary tcpip apps, that being Firefox and Privoxy. Specifically:
1) network.tcp.keepalive.idle_time 600

keepalive values past 120 are really only relevant for dialup, I've used 60 for probably the last 17 years to ensure sockets are never depleted on the server so set the same value on my OS/2 computer with no problems.

As Dave mentioned, ipforward is for using your computer as a router, hence why you need it on for InJoy Firewall or other routing use.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 20, 2018, 10:55:18 am
you are lucky

here setting the same values to my stack IP i have no iprovements at all
still 500-600KiloBytes download from Firefox or from wget at the command prompt on my eCS2.1 desktop with an Intel pro 1000 XT PCI-X Nic… binded at gigabit ethernet to my Cisco gigabit ethernet switch
that communicate with my Cisco 8xx router that runs an FTTC 100Mb/sec…. :(

i've also tried changing NIC card with an Intel pro 100 (100Megabit so, it should be completely sufficient for a download datarate)
with no success at all, here allways max 500-600Kilobytes of download datarates :(

guys, any help?

i've this issue on all my eCS machines :(
the issue is spreading since now users are upgrading internet connections from hyperlan WADSLs (12/1) to FTTCs 50/10 & 100/20 :(


massimo


Ian and Remy,

Guys, another update...I thought I would max out the settings with the goal of achieving highest performance...then if I start running into problems I would dial things back a step at a time.

So here are my current applicable INETCFG.INI settings:

#Inetcfg:       CURRENT DEFAULT MINIMUM MAXIMUM

keepalive       300     300     0       7800    KeepAlive (sec)
tcpswinsize     246723  32768   1024    246723  TCP SendWindow Size
tcprwinsize     246723  32768   1024    246723  TCP RecvWindow Size
udpswinsize     246723  41600   1024    246723  UDP SendWindow Size
udprwinsize     246723  41600   1024    246723  UDP RecvWindow Size
lingertime      120     120     0       65535   Linger Time (sec)
ipforward       0       0       0       1       IP Forwarding flag ON/OFF
timestmp        0       0       0       1       TCP TimeStamps ON/OFF
msl             30      30      1       16384   TCP MaxSegLifeTime (sec)
sack            1       0       0       1       Selective ACK ON/OFF

...the xxxwindsize have been set to MAX values.

I still need to research a bit to understand the impact of the following:

lingertime
ipforward
timestamp (I think this may be used by SACK, not sure though)
msl

....and drummroll for the results please!!!  ;D

1) http://speedsmart.net/ (http://speedsmart.net/) => 87 Mb/s
2) http://testmy.net (http://testmy.net) => 40 Mb/s
3) http://beta.speedtest.net (http://beta.speedtest.net) => 91 Mb/s
4) FF download => 5000 KB/s, so roughly sustained 50 Mb/s

All in all, I would say that is basically what I'm seeing on my Winx boxes.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: David McKenna on June 20, 2018, 11:29:02 pm
Massimo,

  Check inside your C:\TCPIP\BIN\B4TCP.CMD file and make sure it has this:

if exist %etc%\inetcfg.ini CALL inetcfg.exe -set all

 If not add it. This command applies the inetcfg.ini file settings when you boot.

Regards,
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Roderick Klein on June 21, 2018, 01:29:37 am
Ian and Remy,

Guys, another update...I thought I would max out the settings with the goal of achieving highest performance...then if I start running into problems I would dial things back a step at a time.

So here are my current applicable INETCFG.INI settings:

#Inetcfg:       CURRENT DEFAULT MINIMUM MAXIMUM

keepalive       300     300     0       7800    KeepAlive (sec)
tcpswinsize     246723  32768   1024    246723  TCP SendWindow Size
tcprwinsize     246723  32768   1024    246723  TCP RecvWindow Size
udpswinsize     246723  41600   1024    246723  UDP SendWindow Size
udprwinsize     246723  41600   1024    246723  UDP RecvWindow Size
lingertime      120     120     0       65535   Linger Time (sec)
ipforward       0       0       0       1       IP Forwarding flag ON/OFF
timestmp        0       0       0       1       TCP TimeStamps ON/OFF
msl             30      30      1       16384   TCP MaxSegLifeTime (sec)
sack            1       0       0       1       Selective ACK ON/OFF

...the xxxwindsize have been set to MAX values.

I still need to research a bit to understand the impact of the following:

lingertime
ipforward
timestamp (I think this may be used by SACK, not sure though)
msl

....and drummroll for the results please!!!  ;D

1) http://speedsmart.net/ (http://speedsmart.net/) => 87 Mb/s
2) http://testmy.net (http://testmy.net) => 40 Mb/s
3) http://beta.speedtest.net (http://beta.speedtest.net) => 91 Mb/s
4) FF download => 5000 KB/s, so roughly sustained 50 Mb/s

All in all, I would say that is basically what I'm seeing on my Winx boxes.

It makes sense... Windows XP also had IP stack tuners to change some of these value's as you had reduced performance duo to certain IP settings not being tuned automatically.

Roderick
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Roderick Klein on June 21, 2018, 01:32:06 am
Hi Ian!

First of all, big Thank You for putting me on the right path here. The results so far are great, let me explain step at a time below...

...
Manual MTU value depends on what your fibre is using but you are using mtudiscover for MTU so its fine.

You would normally use 1500 for PPPoA, 1492 for PPPoE but it's really depending on your provider. Some will only accept 1492 for PPPoA and 1452 for PPPoE as your on fibre I would expect you to use 1500 and then let mtudiscover look after it. I have no idea if your provider allows jumbo packets or not, I don't even know if OS/2 supports MTU values above 1500 either but normally smaller is better than larger, up to a point...

I re-set the value to 1500 in my \mptn\bin\setup.cmd. My old ISP which utilized PPPoE over DSL line specifically requested 1452, the new guys...well, I did not get any info because the tech doing the install was very tight lipped..LOL, he specifically asked me to call their tech line instead.

...
If you have created an inetcfg.ini file in your x:\mptn\etc directory.

Note you should change your keepalive and sack values to reflect what is below as well.
...

Why do you suggest that SACK be off? I am basiing this on my understanding of that flag...the following article does a pretty nice job of explaining it's functionality => http://packetlife.net/blog/2010/jun/17/tcp-selective-acknowledgments-sack/ (http://packetlife.net/blog/2010/jun/17/tcp-selective-acknowledgments-sack/). In short, if both the client and server support this, when an actual packet is dropped this will allow the re-transmission of ONLY the dropped data, as opposed to a full payload from the dropped point onwards until the end of the conversation. If successful this implies less data being re-sent.

Regarding the keepalive, I will change it to sync it up with my primary tcpip apps, that being Firefox and Privoxy. Specifically:

1) network.tcp.keepalive.idle_time 600

Beyond the above, what remains is to play with the sizes of the values for these key parameters you list below:

...
The main values you want to play with are:-
tcpswinsize
tcprwinsize
udpswinsize
udprwinsize
...

So far here are the results for the following settings:
tcpswinsize     131072
tcprwinsize     131072
udpswinsize     41600
udprwinsize     41600

My FF based downloads are now moving at a speedy 50 Mb/s, a HUGE improvement from where I started!!!

Various internet connection speed test report a steady 90-95 Mb/s rates, this is substantially better as well.

If you want to see your machine go down quicker, do turn on SACK. We experimented with this setting at Mensys on some servers and as somebody else already reported, stay far clear from switching it on. It will cause more pain then its worth...

Roderick
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Neil Waldhauer on June 21, 2018, 07:09:31 am
SpeedSmart gives my speed at 30 MbPS down for my cable modem.

Download from github gives 400 KBPS.

Firefox is the same for both. So Massimo may not have a network problem at all. If he runs SpeedSmart and it shows a good speed, it's likely to be something else. The fact that wget is also slow is probably a clue as to what is really causing slow download.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 21, 2018, 07:27:16 pm
i cheked, that line is there

i've

keepalive 120
tcpswinsize 246723
tcprwinsize  246723
updswinsize  246723
udprwinsize  246723
lingertime 60
ipforward 0
mtudiscover 1
firewall 0

etc..

massimo

Massimo,

  Check inside your C:\TCPIP\BIN\B4TCP.CMD file and make sure it has this:

if exist %etc%\inetcfg.ini CALL inetcfg.exe -set all

 If not add it. This command applies the inetcfg.ini file settings when you boot.

Regards,
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 21, 2018, 08:21:33 pm
that line is there

and my ip stack
has theese values

keepalive 120
all the 4 winsize 246723
lingertime 60
ipforward 0
mtudiscover 1
firewall 0

etc...

massimo

Massimo,

  Check inside your C:\TCPIP\BIN\B4TCP.CMD file and make sure it has this:

if exist %etc%\inetcfg.ini CALL inetcfg.exe -set all

 If not add it. This command applies the inetcfg.ini file settings when you boot.

Regards,
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on June 22, 2018, 03:07:53 pm
Hi Roderick,

If you want to see your machine go down quicker, do turn on SACK. We experimented with this setting at Mensys on some servers and as somebody else already reported, stay far clear from switching it on. It will cause more pain then its worth...

Well, I have been running with SACK turned ON since (at least) the Fibre upgrade, so about 8 months now...no problems to report, at least nothing that I am aware of because it creates such a blatant issue.

However, as you and others have pointed out, the OS/2 tcpip stack apparently has a problem with this...so I'm now in the test mode having turned it OFF.

I will monitor for a while and will provide an update later on. Keep in mind, my extent of running a server process is the Web/2 application, nothing else. I suppose any tcpip apps may be impacted by this though, such as Firefox perhaps? So theoretically I should see something eventually...time will tell.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on June 22, 2018, 03:46:55 pm
Hey maxwarp!

you are lucky

here setting the same values to my stack IP i have no iprovements at all
still 500-600KiloBytes download from Firefox or from wget at the command prompt on my eCS2.1 desktop with an Intel pro 1000 XT PCI-X Nic… binded at gigabit ethernet to my Cisco gigabit ethernet switch
that communicate with my Cisco 8xx router that runs an FTTC 100Mb/sec…. :(
...
guys, any help?
...

Yeah, I feel your pain...when it comes to on-line "feel" nothing beats a fast connection speed and when it's not there it gets pretty frustrating.

OK, so a couple of things:

1) various test servers will be impacted by your local tcpip stack settings, for example, take the SACK flag...as I mentioned in my response to Roderick, I did shut it off right now to test it out, the immediate impact => speedtest.net DIES!!! I get a burst of full speed for about 2-3 secs then the traffic completely stops...judging by this result alone I would be inclined to say that having SACK OFF is a mistake...yet hitting several of the other test servers shows no impact, so your results will vary

2) this is a bit of a follow-up to the idea above, but as applicable to the bigger internet picture...you will get different speeds from different servers, so just because I can do 10 Mbyte to my local test server (same city actually) it does not mean that a server somewhere in China or EU for that matter will show the same results...network latency adds up, etc, etc

3) now to be more specific, can you test what you are seeing using another machine on your LAN? For example, in my testing I was attempting to debug the OS/2 issue but my other boxes (Win7 and mobile devices) were showing decent results...is that the case for you as well? If yes, then indeed, the problem appears to lie with the OS/2 setup itself, if no, then it may very well exist outside of your LAN and on the WAN side

4) there are some more in-depth tools you could use, take a look at the port of Wireshark (https://os2ports.smedley.id.au/index.php?page=wireshark (https://os2ports.smedley.id.au/index.php?page=wireshark)) that Paul Smedley did, it will allow you to visually inspect your tcpip stack traffic and may highlight where your OS/2 specific configuration issue lies, great tunning aid, a bit of a learning curve, you need to capture a tcpip trace on your OS/2 machine then pull this into the Wireshark app, not complex at all, but if you're never done it you just have to try it out a couple of times (look up the thread we had on this forum regarding this, great discussion), heck, that could be a great general discussion of how to tune the OS/2 tcpip stack and maybe deserves a separate thread?
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Roderick Klein on June 22, 2018, 11:00:59 pm

1) various test servers will be impacted by your local tcpip stack settings, for example, take the SACK flag...as I mentioned in my response to Roderick, I did shut it off right now to test it out, the immediate impact => speedtest.net DIES!!! I get a burst of full speed for about 2-3 secs then the traffic completely stops...judging by this result alone I would be inclined to say that having SACK OFF is a mistake...yet hitting several of the other test servers shows no impact, so your results will vary


The advise is based on the first download server that Mensys ever put online for downloads for eCS. I managed 4 servers running OS/2 in a large datacenter with 100 megabit uplink. And with a lot of traffic you could get the TCP/IP stack to get exhaust the system its memory in about 30 minutes by switching SACK on. There is clearly a bug in the stack! Hence the comment on the inetcfg.ini of eCS to *never* turn it on.

Roderick
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 24, 2018, 09:47:09 pm
ok, turned off sack and rebooted the pc
but the download datarate is still like in years 90... :(
upgraded the eCS pc from 2.1 to 2.2b nothing change

i've changed 3 NICs, the intel 1000 embedded on the new mobo Supermicro X7sbe (xeon 4 core)
intel pro100
intel pro1000xt

allways the same slow download datarate
no change :(

massimo


1) various test servers will be impacted by your local tcpip stack settings, for example, take the SACK flag...as I mentioned in my response to Roderick, I did shut it off right now to test it out, the immediate impact => speedtest.net DIES!!! I get a burst of full speed for about 2-3 secs then the traffic completely stops...judging by this result alone I would be inclined to say that having SACK OFF is a mistake...yet hitting several of the other test servers shows no impact, so your results will vary


The advise is based on the first download server that Mensys ever put online for downloads for eCS. I managed 4 servers running OS/2 in a large datacenter with 100 megabit uplink. And with a lot of traffic you could get the TCP/IP stack to get exhaust the system its memory in about 30 minutes by switching SACK on. There is clearly a bug in the stack! Hence the comment on the inetcfg.ini of eCS to *never* turn it on.

Roderick
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 24, 2018, 09:56:56 pm
yes, it's all there

each time i modify a parameter i do a reboot of the pc
after the reboot i also do inetcfg -get all
to check


massimo

Massimo,

  Check inside your C:\TCPIP\BIN\B4TCP.CMD file and make sure it has this:

if exist %etc%\inetcfg.ini CALL inetcfg.exe -set all

 If not add it. This command applies the inetcfg.ini file settings when you boot.

Regards,
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 24, 2018, 10:00:21 pm
no, since all my eCS 2.1 or 2.2b desktop/servers/firewalls give a false/erroneus downstreaming result

about 13Megabit when connections are FTTCs at 50 or 100Megabits
so it's eCS that have some issue

different hardwares even xeon 4 cores, and all runs at 13Megabits, the fast download datarate i get is 2MegaBytes/sec

while W10 on the same lan even with an old asus eee pc dual atom with 100Megabit nic download at full data rate

even W10 PCs under theese eCS firewalls download at 13Megabit/sec erroneus speed rate

speed is there, but eCS 2.x tcpip stack slow down it to a maximum of 2MegaBytes/sec

massimo
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 24, 2018, 10:08:35 pm
of course i really hope that i can fix this embarassing situation :(
since elsewhere i will get in serious troubles since all my firewalls are eCS+Injoy

i'm going to buy an ArcaOS to see if fix this issue


massimo
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on June 24, 2018, 11:47:05 pm
Interestingly, with the default ArcaOS default settings and a NVidia MCP51 controller on a LTE network, I usually get about 12 Mb/s down. Tried in the morning when I guess the network wasn't so congested and got 25 Mb/s, which sounds better then your fibre connection. All the machines connected seem to get the same no matter what OS is running.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Daniel Carroll on June 25, 2018, 01:05:53 am
I will certainly be interested in any information discovered. My ISP, Sonic Telecom, just emailed that they will be providing fiber to my home in September and for the same amount they have been charging for DSL I will get 1Gb/s (symmetrical). Woohoo!  :)
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 25, 2018, 09:46:21 am
about wireshark

i've tried to download wireshark (had to use my w10 laptop since unable to download from eCS'firefox), but as usual install something under eCS is a lot frustrating
i've all GCC/KLIBC DLLs updated, but it complains that it need a dll "rocken.dll"
searching on 3 different search engines i've found anything (also dunno why eCS devs hate search engines that are not the shi$$y googl monopolist)

that dll is not mentioned neither in the docs of the software

guys, we live in the era of terabytes hard-disks, we don't have anymore to deal with 100MegaBytes disk partitions, dunno why eCS devs still don't
bundle all the needed DLLs in a subdir in the package :-(
that's a lot frustrating :(




Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on June 25, 2018, 03:05:56 pm
massiomo,

about wireshark
...
i've all GCC/KLIBC DLLs updated, but it complains that it need a dll "rocken.dll"
...
that dll is not mentioned neither in the docs of the software
...

I took a closer look at this, this DLL does not appear to be Wireshark related, for one thing, it is not present on my machine and I am able to successfuly use Wireshark.

My guess is that something else on your system appears to need this.

Do you have the 'DLL Tree' utility? This allows you to test load an EXE (or DLLs) and will map out (visually) what other DLLs are needed. I would suggest that you run wireshark.exe through that app and see where rocken.dll is being requested.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: ivan on June 25, 2018, 05:22:50 pm
Hi Dariusz,

I assume you mean PMDLL-2.10 when you mentioned the 'DLL Tree' utility.  It is on hobbes but to make it easy I've attached it here.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on June 25, 2018, 06:45:28 pm
I assume you mean PMDLL-2.10 when you mentioned the 'DLL Tree' utility...

Ivan,
Yes, that is correct...
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Greg Pringle on June 25, 2018, 11:18:01 pm
This might sound academic but have you changed the cables? I have found some cables will not work at high speeds. Change to cat Cat 6 between your computer and router and see if it performs better.

Here is a speed test I just did between an ArcaOS machine and an Ecomstation machine on the local net with the Ecomstation machine on the other side of a local router. (So the trafic is routed instead of switched).

ftp> bin
200 Type set to I.
ftp> put efs097v11a4_all.tar
200 PORT command successful.
150 Opening Binary mode data connection for /tmp/efs097v11a4_all.tar
Sent 13178880 bytes
226 Binary transfer complete.
local: efs097v11a4_all.tar remote: efs097v11a4_all.tar
13178880 bytes sent in 1.36 seconds (9463 Kbytes/s)

Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on June 26, 2018, 02:15:59 pm
hey, all CAT7 SFTP here, the switch is a gigabit ethernet layer3 cisco sg300-28  :)

NIC is binded at 1Gigabit full duplex

anyway doing ftp from other machines in my lan i get about 22MegaBytes/sec
the issue is only with downloading from the internet
where all W10 machines are a very lot faster, even the very old netbook asus eee pc and the smartphone :(

i've tried right now to download a big file from a eCS2.1 server in my LAN and it download via the old ibm ftp at 13MegaBytes/sec
that is not excellent, but surely quite good and sufficient for me

i repeat the issue is downloading http/s or ftp from the internet only
of course i see performances not real (in downstreaming) also from the varius speedtests (eg. ookla speedtest.net)

massimo
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Rich Walsh on June 27, 2018, 06:42:06 am
i repeat the issue is downloading http/s or ftp from the internet only

Since no one else has mentioned these:

MTU - are you absolutely sure each system's MTU  is set correctly for your WAN? I was getting half-speed out of my connection until I realized MTU was set to the default of 1500 while my ISP's network needed 1492. Once that was corrected, I started getting the same speed I get with Win or Lin.

Injoy Firewall - I know almost nothing about this product but recall seeing a recent bug report somewhere claiming a major slow-down when it was active. Regardless, a direct connection to your provider's box by a machine that isn't weighed-down by IP-addons like a software firewall might be informative.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on June 27, 2018, 08:07:17 am
Hi!

I use Injoy (latest build) too and I could do test without it (same results) as well checked and set the correct maximum MTU.
I have a similar problem, speed didn't go above 15Mbps while I have 75Mbps under windows.

Regard,
Rémy

(older network specialist)
 
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on June 27, 2018, 08:37:48 am
i repeat the issue is downloading http/s or ftp from the internet only

Since no one else has mentioned these:

MTU - are you absolutely sure each system's MTU  is set correctly for your WAN? I was getting half-speed out of my connection until I realized MTU was set to the default of 1500 while my ISP's network needed 1492. Once that was corrected, I started getting the same speed I get with Win or Lin.

I was going to mention this, a problem that was particularly bad on eCS as they had mtudiscover mis-configured to not discover the mtu. Worth trying to force a small number such as 576 and working up. Might have to set mtudiscover to zero before testing with something like "ifconfig lan0 mtu 576'
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Greg Pringle on June 27, 2018, 03:49:57 pm
There is an easier way to find the actual MTU size using OS/2.

From a command prompt run ping like this:
ping www.yahoo.com 1472 2
PING atsv2-fp.wg1.b.yahoo.com: 1472 data bytes
1480 bytes from 72.30.35.10: icmp_seq=0. time=20. ms
1480 bytes from 72.30.35.10: icmp_seq=1. time=20. ms

Shown are the actual results.
If this shows results then do it again and increase the number 1472 until the ping does not return results.

If this does not return then lower the number 1472 until the ping returns values.

Once the highest number is found that returns results then add 28 to that for the MTU size.

So, in the above 1472 + 28 = 1500
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on June 28, 2018, 05:50:05 am
Interesting, didn't realize that our ping could do that. do get weird result though,
Code: [Select]
E:\test>ping www.yahoo.com 996 2
PING atsv2-fp.wg1.b.yahoo.com: 996 data bytes
1004 bytes from 98.137.246.8: icmp_seq=0. time=50. ms
1004 bytes from 98.137.246.8: icmp_seq=1. time=60. ms

----atsv2-fp.wg1.b.yahoo.com PING Statistics----
2 packets transmitted, 2 packets received, 0% packet loss
round-trip (ms)  min/avg/max = 50/55/60

E:\test>ping www.yahoo.com 997 2
PING atsv2-fp.wg1.b.yahoo.com: 997 data bytes

----atsv2-fp.wg1.b.yahoo.com PING Statistics----
2 packets transmitted, 0 packets received, 100% packet loss

Which implies that my MTU should be 1024, yet things seem to work fine with it at 1500.
Testing with SpeedSmart, the numbers are all over the place no matter which MTU I use, have to test when the neighbours aren't watching Netflix :)
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Lars on June 28, 2018, 08:12:56 am
For everyone using the (lousy) OS/2 DHCP client (dynamic IP address served by a DHCP server instead of static address), you need to add this line to %ETC%\dhcpcd.cfg:

ifconfig_options "mtu 1492 -trailers"

Adjust to your needs. "-trailers" promises some speed improvement, I don't know if that helps.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on June 28, 2018, 06:10:06 pm
If you are using static IP's, put it in your x:\mptn\bin\setup.cmd file on the ipconfig line for your interface, ie

    ifconfig lan0 192.168.1.4 netmask 255.255.255.0 -trailers

trailers have been off for linux/unix for a long time, though there it is specified as 'notrailers', it applies to static as well as dynamic IP's.
If you are only using dhcp, you can add it as "ifconfig lan0 -trailers" to your setup.cmd file or as you say, to your dhcpcd.cfg file.

I looked at it decade(s) ago and could not see any difference. YMMV

For OS/2, '-trailers'  means turn it off, 'trailers' means turn it on. From memory that's how it worked.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on June 29, 2018, 01:55:27 am
According to the TCPIP documentation on ifconfig, -trailers is the default. For DHCP, it is less clear,but to quote the DHCP server documentation,
Quote
This option is specified only at the DHCP server.

Client negotiates (1) or does not negotiate (0) the use of trailers when using Address Resolution Protocol (ARP).   For more information, see RFC 893.

Configuration file format: Boolean

So not much point if adding it to the clients DHCPCD.CFG file and the templates for the server default to off. Now to read RFC 893, http://www.faqs.org/rfcs/rfc893.html (http://www.faqs.org/rfcs/rfc893.html)
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Andy Willis on June 29, 2018, 05:47:28 am
For everyone using the (lousy) OS/2 DHCP client (dynamic IP address served by a DHCP server instead of static address), you need to add this line to %ETC%\dhcpcd.cfg:

ifconfig_options "mtu 1492 -trailers"

Adjust to your needs. "-trailers" promises some speed improvement, I don't know if that helps.

Is this equivalent to:
option 34 on
or
option 34 off
?
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on June 29, 2018, 05:56:21 am
For everyone using the (lousy) OS/2 DHCP client (dynamic IP address served by a DHCP server instead of static address), you need to add this line to %ETC%\dhcpcd.cfg:

ifconfig_options "mtu 1492 -trailers"

Adjust to your needs. "-trailers" promises some speed improvement, I don't know if that helps.

Off.
Quote
Is this equivalent to:
option 34 on
or
option 34 off
?
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on June 29, 2018, 01:19:40 pm
For everyone using the (lousy) OS/2 DHCP client (dynamic IP address served by a DHCP server instead of static address), you need to add this line to %ETC%\dhcpcd.cfg:

ifconfig_options "mtu 1492 -trailers"

Adjust to your needs. "-trailers" promises some speed improvement, I don't know if that helps.

Does not improve performance here.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on July 12, 2018, 08:22:13 am
Found this for testing,
Code: [Select]
wget -O speedtest-cli https://raw.githubusercontent.com/sivel/speedtest-cli/master/speedtest.py
python speedtest-cli
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: TeLLie on July 12, 2018, 06:29:08 pm
I have a 200 Mbit/s connection
And i get this:
Retrieving speedtest.net configuration...
Testing from Xs4all Internet BV (83.161.157.109)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Tilaa B.V. (Amsterdam) [6.94 km]: 5.833 ms
Testing download speed...............................
Download: 208.87 Mbit/s
Testing upload speed.................................
Upload: 176.00 Mbit/s

I use ArcaNoa
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on July 15, 2018, 03:03:51 pm
Here are my results on a 125Mbit line:

Code: [Select]
[G:\tmp]python speedtest-cli
Retrieving speedtest.net configuration...
Testing from Cogeco Cable (24.57.68.234)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by MNSi Telecom (Windsor, ON) [3.24 km]: 38.167 ms
Testing download speed..........................................................
......................
Download: 109.61 Mbit/s
Testing upload speed............................................................
....................................
Upload: 6.79 Mbit/s
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on July 16, 2018, 02:21:54 pm
I did new tests... (use of DHCP in both test from default install)

Booting ArcaOS disk -> speed test nearly 10Mbps

I then did a test installing ArcaOS into Virtualbox (using default "NAT mode" Virtualbox network driver then the native installation one without virtualbox) under windows 7 -> speed test gives nearly 58Mbps !!

For me it looks the native driver used with my onboard network card (Realtek® 8111G Gigabit LAN) is not well working and is the culprit of bad speed. The driver has to be corrected and is buggy and this test confirms other tests I already did. 
 
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Roderick Klein on July 16, 2018, 05:23:48 pm
I did new tests... (use of DHCP in both test from default install)

Booting ArcaOS disk -> speed test nearly 10Mbps

I then did a test installing ArcaOS into Virtualbox (using default "NAT mode" Virtualbox network driver then the native installation one without virtualbox) under windows 7 -> speed test gives nearly 58Mbps !!

For me it looks the native driver used with my onboard network card (Realtek® 8111G Gigabit LAN) is not well working and is the culprit of bad speed. The driver has to be corrected and is buggy and this test confirms other tests I already did. 
 

But how do you know this is NOT the IP stack of OS/2 ?

* With my Intel E1000B.OS2 driver I get about 25 megabyte per second over my local LAN to a Linux server.

* Speedtest.net indicates a 100 megabit and 67 megabit upload speed via the internet.

* I have a website I only get 350 KB per second from. Windows delivers 6.4 megabyte per second to the same website. Now here in the interesting, I repeated the test and hooked up a ethernet  to wifi adapter and setup my mobile phone as wifi hotspot to connect to the same website. Guess what its faster then 350 KB second!
So purely changing the data path shows the issue can also have todo with the IP stack somehow. Tracing that is part 2!

I am currently looking at IP trace and wireshark to see if something is going wrong with packet processing!
Like re-transmits or packets out of sync.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on July 16, 2018, 10:13:36 pm
You can give a try to this "speedtest' too
https://openspeedtest.com/Get-widget.php
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Roderick Klein on July 17, 2018, 12:14:26 am
You can give a try to this "speedtest' too
https://openspeedtest.com/Get-widget.php

I get very low upload of only 2 megabit. That said I have no clue where this server is located. Also take into account that you also need to know where your download test server is located on the globe. In this case the speedtest.net I use is about 20 KM away. So does test resulsts do not say me anything.

Do they have local server around the globe or just one server ?
Running an IP trace I get 18 hopes... I get send over to Texas to the IP adress 45.79.12.7.
But this is not even the end of the trace route. Anyay for a European user a pretty bad example to use for speedtest in my opinion.

Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on July 17, 2018, 01:43:08 am
speedtest-cli will find the closest server, based on ping. Not the most accurate as it claims the server is only 1.55 km away, when Vancouver is 80 kms away, perhaps 100km by fibre.
Quote
E:\test>python speedtest-cli
Retrieving speedtest.net configuration...
Testing from Telus Communications (209.52.88.204)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Skyway West Business Internet Services (Vancouver, BC) [1.55 km]: 56.0
 ms
Testing download speed..........................................................
......................
Download: 22.92 Mbit/s
Testing upload speed............................................................
....................................
Upload: 2.20 Mbit/s

Not too bad for wireless LTE connection. Gets much slower in the evening when everyone is watching netflix etc.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on July 17, 2018, 01:45:58 am
@Remy. Which driver are you using? It might be worth trying some alternatives and/or opening an issue at Arca Noae. I'm currently using a Realtek® 8111 Gigabit LAN or such and get not bad speeds with the mmre driver but I understand Realtek changes things often and sometimes different drivers are better.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on July 17, 2018, 09:01:00 am
Installed driver:
MMRE RealTek 8168/8111 B/C/CP/D/DP/E/F/G PCIe Gigabit Ethernet Network Adapter
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on July 17, 2018, 03:55:51 pm
I changed the driver for "Realtek 8169/1Gb PCIe MAC Driver"
Now, I have 25Mbps which nearly 2 times faster than the MMRE driver but always slower than under windows or using the network driver for under virtualbox.

Updt: above tests where done from under firefox.
(With the MMRE driver, I didn't see any noticeable speed difference from firefox test or under an os/2 window)

I just  tested the ""Realtek 8169/1Gb PCIe MAC Driver"" from under an OS/2 window and this driver is definitively the good one to use because I could get a 56Mbps download speed and 5Mbps upload which is now a good value due having just do a test under windows and currently, my download speed is nearly 58Mbps (one of the lowest speed I had using the same hosted speedtest server)

===================================================
[C:\speedtest]python speedtest-cli
Retrieving speedtest.net configuration...
Testing from Bouygues Telecom .......
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by ONLINE S.A.S. (xxxxxxxxxxxxxx) [20.07 km]: 16.333 ms
Testing download speed.......................
Download: 56.11 Mbit/s
Testing upload speed............................
Upload: 5.10 Mbit/s
===================================================


From under firefox (the one under ArcaOS 38, the speed didn't go above 25Mbps !
Firefox looks to be a bottleneck now...

(I tested again the speed using the MMRE driver and get 9.8Mbps !)

Thanks having suggested me to use the other available driver I didn't try.   
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on July 19, 2018, 01:47:56 am
I just installed firefox 45.9 and the speed botleneck from ff38 seems to be resolved.
Under ff45.9, the speed test result is very simlar to speetest cli os/2 prompt cmd.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on July 20, 2018, 07:09:55 pm
Hi,

I did more test during my last updates and found an interresting thing...
When putting the MMRE driver before the FXWRAP (injoy firewall) driver, the speed increases a lot.
For those using IJFW, I suggest them having the FXWRAP device after the network device like :
DEVICE=C:\IBMCOM\MACS\MMRE.OS2
DEVICE=C:\IBMCOM\PROTOCOL\FXWRAP.SYS

After doing this, the speed is mutch better (34Mbps) and stopping the IJFW server, the speed goes up to 56Mbps now...   
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on July 21, 2018, 06:24:19 am
Hi Remy,

Quote
When putting the MMRE driver before the FXWRAP (injoy firewall) driver, the speed increases a lot.
For those using IJFW, I suggest them having the FXWRAP device after the network device

The FXWRAP driver is ment to come after the NIC driver to the outside/external/unprotected enviroment but before the NIC driver to the inside/internal/protected network, otherwise the firewall cannot do its job 100%.

example:

DEVICE=C:\IBMCOM\MACS\MMALC.OS2
DEVICE=C:\IBMCOM\PROTOCOL\FXWRAP.SYS /S
DEVICE=C:\IBMCOM\MACS\E1000.OS2

MMALC.OS2 is for the NIC connected to my external modem/router
FXWRAP.SYS is the InJoy firewall driver that will intercept the datastream from MMALC.OS2 and establish itself before the driver the the internal network is loaded.
E1000.OS2 is for the NIC connected to my internal network.

Added: Putting another driver with software between interfaces will also slow down a system depending on how much grunt your system has and the number of firewall rules. Just something we live with for the added protection.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Rene Hvidsoe on July 21, 2018, 11:58:41 am
Hi,

I have been following the thread. I am using the MMRE driver for my RTL 8110 xE fast express Ethernet (device:8136h). I have for a few years ago also slow upload/download. I have at that time just bought Injoy Firewall and running Ecomstation 2.1 and could not figure out why my net performance was poor compared to Windows and Linux on same PC. I reported it to Bitwise and it turns out that Firewall was the issue. I needed to adjust the  firewall priority to 100. Since then I have the same performance on all OS Windows/Linux and ArcaOS.

I am using latest MMRE driver.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on July 21, 2018, 02:47:16 pm
Hi,

I have been following the thread. I am using the MMRE driver for my RTL 8110 xE fast express Ethernet (device:8136h). I have for a few years ago also slow upload/download. I have at that time just bought Injoy Firewall and running Ecomstation 2.1 and could not figure out why my net performance was poor compared to Windows and Linux on same PC. I reported it to Bitwise and it turns out that Firewall was the issue. I needed to adjust the  firewall priority to 100. Since then I have the same performance on all OS Windows/Linux and ArcaOS.

I am using latest MMRE driver.

Thanks for your advice, I never tried to put priority to 100 due it is writtent it could have bad effect...
I'll give it a try.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on July 21, 2018, 03:38:16 pm
I did the test, setting ijfw to priority 100 didn't provide my any speed gain : same as priority 40
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Julien Pierre on January 11, 2019, 01:37:29 am
I have a gigabit connection , with 40 mbps upload speed.

With speedtest-cli, on a Windows host, I get this :
Retrieving speedtest.net configuration...
Testing from Comcast Cable (67.169.190.100)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Sonic.net, Inc. (San Jose, CA) [8.57 km]: 14.29 ms
Testing download speed................................................................................
Download: 377.07 Mbit/s
Testing upload speed...
Upload: 40.74 Mbit/s

Whereas I get upwards of 900 Mbps in Firefox 64.0.2 with speedtest.net on the same Win10 machine.
speedtest-cli ini Python has a bottleneck around 400 Mbps. I get the same limit on an ODROID XU4 .

Now, with ArcaOS 5.0.3 running in a VM on the same Win10 host, in Firefox 38, I get :
56 Mbps down / 26 Mbps up .


And with speedtest-cli :

Retrieving speedtest.net configuration...
Testing from Comcast Cable (67.169.190.100)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Sonic.net, Inc. (San Jose, CA) [8.57 km]: 18.167 ms
Testing download speed..........................................................
......................
Download: 124.07 Mbit/s
Testing upload speed............................................................
....................................
Upload: 36.23 Mbit/s

Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Julien Pierre on January 11, 2019, 01:40:36 am
I just installed firefox 45.9 and the speed botleneck from ff38 seems to be resolved.
Under ff45.9, the speed test result is very simlar to speetest cli os/2 prompt cmd.

Where did you obtain this "new" FF 45.9 ?
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on January 11, 2019, 01:49:10 am
My builds are here, https://bitbucket.org/dryeo/dry-comm-esr31/downloads/ (https://bitbucket.org/dryeo/dry-comm-esr31/downloads/) seem a bit more stable then the official builds, probably due to newer compiler, also support SSE2 which the official one doesn't due to compiler bug. Note the Sandy bridge one won't run on the regular OS/2 kernel due to lack of AVX support. Need the OS4 kernel.
The official is at github, https://github.com/bitwiseworks/mozilla-os2/releases (https://github.com/bitwiseworks/mozilla-os2/releases), there are also RPM packaged versions around if you have access to the repositories.
edit, need to add libvpx and the hunspell spelling checker rpms, FFmpeg libs are optional but needed for h264 playback.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Julien Pierre on January 11, 2019, 01:52:29 am
My builds are here, https://bitbucket.org/dryeo/dry-comm-esr31/downloads/ (https://bitbucket.org/dryeo/dry-comm-esr31/downloads/) seem a bit more stable then the official builds, probably due to newer compiler, also support SSE2 which the official one doesn't due to compiler bug. Note the Sandy bridge one won't run on the regular OS/2 kernel due to lack of AVX support. Need the OS4 kernel.
The official is at github, https://github.com/bitwiseworks/mozilla-os2/releases (https://github.com/bitwiseworks/mozilla-os2/releases), there are also RPM packaged versions around if you have access to the repositories.

Thanks. I just got it from the RPM. Unfortunately, it did not improve my download speed at all in Firefox with speedtest.net . This is in a VM, not physical machine.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on January 11, 2019, 02:00:26 am
Quite possibly a VM needs different tuning then a real machine. As discussed, the default TCP/IP tuning isn't the best on our ancient OS.
You mentioned eCS, make sure you change mtudiscover to the default of 1 in %ETC%\inetcfg.ini.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on January 11, 2019, 03:21:57 am
Hi Julien,

I have a gigabit connection , with 40 mbps upload speed.

With speedtest-cli, on a Windows host, I get this :
Retrieving speedtest.net configuration...
Testing from Comcast Cable (67.169.190.100)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Sonic.net, Inc. (San Jose, CA) [8.57 km]: 14.29 ms
Testing download speed................................................................................
Download: 377.07 Mbit/s
Testing upload speed...
Upload: 40.74 Mbit/s

Whereas I get upwards of 900 Mbps in Firefox 64.0.2 with speedtest.net on the same Win10 machine.
speedtest-cli ini Python has a bottleneck around 400 Mbps. I get the same limit on an ODROID XU4 .
...

Well, having done a bit more work with my setup here, and in particular comparing it to the other machines, including one pretty darn top-line Win10 box (Ryzen 7 2700x) there indeed are huge difference in performance. But...I attribute a lot of that to the hardware itself.

For example, my current OS/2 box shows:

Code: [Select]
Retrieving speedtest.net configuration...
Testing from Bell Canada (50.100.198.117)...
Retrieving speedtest.net server list...
Retrieving information for the selected server...
Hosted by TekSavvy Solutions Inc (Toronto, ON) [332.79 km]: 29.667 ms
Testing download speed..........................................................
......................
Download: 350.98 Mbit/s
Testing upload speed............................................................
....................................
Upload: 527.68 Mbit/s

Consistently I see my u/l speed being faster than the d/l part of it, and I have seen numerous d/l speeds over 400Mb/s while my u/l almost never drop to below 500, so at least here the Python environment allows me to plug ahead.

Now my Win10 box blow this away with a d/l of 850-900 and u/l of 930-950. I have not hit that 960 max that you really might consider to be a ceiling due to overheads, etc. etc.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Julien Pierre on January 11, 2019, 10:51:56 pm
Quite possibly a VM needs different tuning then a real machine. As discussed, the default TCP/IP tuning isn't the best on our ancient OS.
You mentioned eCS, make sure you change mtudiscover to the default of 1 in %ETC%\inetcfg.ini.

I have both ECS 2.0 and ArcaOS 5.0.3, but they are each running in a VM.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Julien Pierre on January 11, 2019, 10:54:38 pm
Code: [Select]
[quote author=Dariusz Piatkowski link=topic=1486.msg20189#msg20189 date=1547173317]
Retrieving speedtest.net configuration...
Testing from Bell Canada (50.100.198.117)...
Retrieving speedtest.net server list...
Retrieving information for the selected server...
Hosted by TekSavvy Solutions Inc (Toronto, ON) [332.79 km]: 29.667 ms
Testing download speed..........................................................
......................
Download: 350.98 Mbit/s
Testing upload speed............................................................
....................................
Upload: 527.68 Mbit/s

Consistently I see my u/l speed being faster than the d/l part of it, and I have seen numerous d/l speeds over 400Mb/s while my u/l almost never drop to below 500, so at least here the Python environment allows me to plug ahead.
[/quote]

Thanks, that's interesting. Comcast here in the US considers fast upload speed to be business class, and charges an arm and a leg for them. I have heard >$400/month for fiber. I don't think it's available at my house, not that it matters. Anyway, I'm stuck with 40 Mbps upload speed. And a 1TB/month data cap. Sigh !

Quote
Now my Win10 box blow this away with a d/l of 850-900 and u/l of 930-950. I have not hit that 960 max that you really might consider to be a ceiling due to overheads, etc. etc.

My Win10 box is a 5820k overclocked at 4.3 Ghz . I would have expected it to get closer in speedtest-cli. It does get to 900+ in Firefox, but not in speedtest-cli . Interesting that you are able to beat that on the Ryzen.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Lars on January 12, 2019, 02:33:41 pm
Hi Remy,

Quote
When putting the MMRE driver before the FXWRAP (injoy firewall) driver, the speed increases a lot.
For those using IJFW, I suggest them having the FXWRAP device after the network device

The FXWRAP driver is ment to come after the NIC driver to the outside/external/unprotected enviroment but before the NIC driver to the inside/internal/protected network, otherwise the firewall cannot do its job 100%.

example:

DEVICE=C:\IBMCOM\MACS\MMALC.OS2
DEVICE=C:\IBMCOM\PROTOCOL\FXWRAP.SYS /S
DEVICE=C:\IBMCOM\MACS\E1000.OS2

MMALC.OS2 is for the NIC connected to my external modem/router
FXWRAP.SYS is the InJoy firewall driver that will intercept the datastream from MMALC.OS2 and establish itself before the driver the the internal network is loaded.
E1000.OS2 is for the NIC connected to my internal network.


That should not matter. The PROTOCOL.INI defines what protocol driver links to what MAC driver. And is is the job of NETBIND.EXE to make the protocol driver known to the MAC driver and vice versa so that these can exchange data. In order to do that it uses PROTOCOL.INI. And NETBIND.EXE runs long after the protocol and MAC drivers are loaded via config.sys.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Julien Pierre on January 12, 2019, 10:07:04 pm
Hi Remy,

Quote
When putting the MMRE driver before the FXWRAP (injoy firewall) driver, the speed increases a lot.
For those using IJFW, I suggest them having the FXWRAP device after the network device

The FXWRAP driver is ment to come after the NIC driver to the outside/external/unprotected enviroment but before the NIC driver to the inside/internal/protected network, otherwise the firewall cannot do its job 100%.

example:

DEVICE=C:\IBMCOM\MACS\MMALC.OS2
DEVICE=C:\IBMCOM\PROTOCOL\FXWRAP.SYS /S
DEVICE=C:\IBMCOM\MACS\E1000.OS2

MMALC.OS2 is for the NIC connected to my external modem/router
FXWRAP.SYS is the InJoy firewall driver that will intercept the datastream from MMALC.OS2 and establish itself before the driver the the internal network is loaded.
E1000.OS2 is for the NIC connected to my internal network.


That should not matter. The PROTOCOL.INI defines what protocol driver links to what MAC driver. And is is the job of NETBIND.EXE to make the protocol driver known to the MAC driver and vice versa so that these can exchange data. In order to do that it uses PROTOCOL.INI. And NETBIND.EXE runs long after the protocol and MAC drivers are loaded via config.sys.

FYI, I don't have FXWRAP in my config.sys .
I would appreciate any info on tuning ArcaOS for a VM.
I tried both E1000 and AMD PCnet interfaces in Virtualbox, but didn't see any difference with speedtest-cli .

I still think a native code benchmark would be preferable to one written in Python.

For testing LAN performance, there is iperf3 on other platforms, but I have not found an OS/2 port.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on January 13, 2019, 02:27:54 am
I still think a native code benchmark would be preferable to one written in Python.

For testing LAN performance, there is iperf3 on other platforms, but I have not found an OS/2 port.

There is https://88watts.net/download/netio-1.33.zip (https://88watts.net/download/netio-1.33.zip) for testing LAN performance. http://hobbes.nmsu.edu/download/pub/os2/util/network/netio132.zip (http://hobbes.nmsu.edu/download/pub/os2/util/network/netio132.zip) has binaries for most other systems.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Julien Pierre on January 13, 2019, 03:08:09 am
I still think a native code benchmark would be preferable to one written in Python.

For testing LAN performance, there is iperf3 on other platforms, but I have not found an OS/2 port.

There is https://88watts.net/download/netio-1.33.zip (https://88watts.net/download/netio-1.33.zip) for testing LAN performance. http://hobbes.nmsu.edu/download/pub/os2/util/network/netio132.zip (http://hobbes.nmsu.edu/download/pub/os2/util/network/netio132.zip) has binaries for most other systems.

Thanks, I'll take a look. I now have a 10 Gigabit ethernet LAN. With iperf3, I can get 9.9 Gbps with 2 streams using jumbo frames between my Ubuntu NAS and my Win10 desktop on a Netgear GS110MX switch. With real file transfers, I get about 3 to 5 Gbps to and from a 50TB RAIDZ2 array. I will check first if netio is up to the 10 Gbps challenge between Linux and Win10 first, before I try it on ArcaOS.

I used a quick file system benchmark for OS/2 that I wrote 25 years ago. Ran it in my ArcaOS VM . I mounted shares from my Win10 host (local) and Ubuntu AS (remote). With ArcaMapper.

On large files (multi GB), I got 34MB/s average from the NAS to VM. And 54MB/s from a share on the Win10 host to ArcaOS guest. Those are only about 272 and 432 Mbps respectively. I should be able to at least get Gigabit speed on the LAN and local share, and hopefully more, but it seems the networking in Virtualbox is just too slow. I might get faster results with iperf and/or netio as it would take the SMB out of the picture.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Julien Pierre on January 13, 2019, 04:42:07 am
On large files (multi GB), I got 34MB/s average from the NAS to VM. And 54MB/s from a share on the Win10 host to ArcaOS guest. Those are only about 272 and 432 Mbps respectively. I should be able to at least get Gigabit speed on the LAN and local share, and hopefully more, but it seems the networking in Virtualbox is just too slow. I might get faster results with iperf and/or netio as it would take the SMB out of the picture.

So, I broke the Gigabit limit on my LAN in my OS/2 VM.

Here is how I did it :
1) upgraded to Virtualbox 6.0
2) noticed there were new additions with a VBOXFS.IFS for shared folders
3) shared my X: drive from my Win10 host to ArcaOS guest as X: also .
X: is a share mounted from my Ubuntu NAS to the Win10 host.
4) ran my file benchmark again, on a 4.5GB ISO movie file
I averaged 243,677 KB/s , which is 1949 Mbps , or about 2 Gbps. This is consistent with what I was seeing in the Win10 host task manager - network bandwidth at 2 Gbps .
Subsequent runs are even faster, because the Win10 host is caching the remote file : 608,601 KB/s . This is about 5 Gbps . In this case there as 0 network bandwidth usage, though, due to caching.

That means I have to try with files that don't fit in the host's RAM (32GB), or just try with different (uncached) files each time.

I didn't try yet sharing a local drive to the VM yet. I have a striped array of SSDs which does 2600MBps sequential (about 21Gbps) in Crystaldiskmark 6 for sequential test. My guess is that I will hit a bottleneck much lower than that in the VM.

Just tried it, since adding shared folders don't require a VM restart.
I was able to read a 35GB file, uncached, at 656,854 KB/s from the local SSD array. I tried again with a 2GB file, twice to make sure it was cached, and got similar speed. So, 5 Gbps appears to be the upper limit for the shared folders in Virtualbox.


Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on September 07, 2019, 03:44:41 pm
Just to add some found information:

- A good site to test bandwide  https://www.nperf.com  the best and fastest available result
- To well tune RWIN => https://www.speedguide.net    (the given values works and if a multiple value is used as suggested, I have less slowdown.
   In my case, I set the RWIN to 128480  because the max allowed into inetcfg.ini is not a multiple of MSS (MSS=1460 and MTU=1500 in my case)
   
I have a speed near the maximum allowed for a 100Mb connexion
The remaining problem I have is that the interface did not recognize/enable Gigabit link like windows do and refuses to go above 100Mb.
I tried to force it to be Gigabit but failled until now. 

regards 
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on September 07, 2019, 11:59:33 pm
Hi Remy, everyone...

...In my case, I set the RWIN to 128480  because the max allowed into inetcfg.ini is not a multiple of MSS (MSS=1460 and MTU=1500 in my case)...

I've played around with these a few times, so far this is my current best-combo:

Code: [Select]
#Inetcfg: CURRENT DEFAULT MINIMUM MAXIMUM

keepalive 60 7800 0 7800 KeepAlive (sec)
tcpswinsize 246723 32768 1024 246723 TCP SendWindow Size
tcprwinsize 246723 32768 1024 246723 TCP RecvWindow Size
udpswinsize 246723 9216 1024 246723 UDP SendWindow Size
udprwinsize 246723 41600 1024 246723 UDP RecvWindow Size
lingertime 120 120 0 65535 Linger Time (sec)
probecnt 8 8 1 8 Number Of KeepAlive Probes
ipforward 0 0 0 1 IP Forwarding flag ON/OFF
mtudiscover 1 1 0 1 TCP Path MTU Discovery ON/OFF
arpkillc 1200 1200 180 15300 ARP Complete TimeOut (sec)
arpkilli 180 180 60 1200 ARP InComplete TimeOut (sec)
lipcstswinsize 246723 10240 1024 246723 LIPC Stream SendWindow Size
lipcstrwinsize 246723 10240 1024 246723 LIPC Stream RecvWindow Size
lipcdgswinsize 246723 2048 1024 246723 LIPC Dgram SendWindow Size
lipcdgrwinsize 246723 4096 1024 246723 LIPC Dgram RecvWindow Size
synattack 0 0 0 1 SYN Attack Flag ON/OFF
syncookie 0 0 0 1 SYN Cookie Flag ON/OFF
firewall 0 0 0 1 FIREWALL ON/OFF
multidefrt 1 1 0 1 Multiple Default Routes ON/OFF
winscale 1 1 0 1 TCP Window Scale ON/OFF
timestmp 0 0 0 1 TCP TimeStamps ON/OFF
realslow 20 20 1 100 TCP TW Q Slow Timeout ticks
perfhttp 1 1 0 1 Fast Path HTTP ON/OFF
tcpttl 64 64 1 255 TCP  TTL (hops)
udpttl 64 64 1 255 UDP  TTL (hops)
icmpttl 64 64 1 255 ICMP TTL (hops)
fragttl 60 60 1 255 Fragment TTL (sec)
reusetw 1 1 0 1 Reuse Time Wait ON/OFF
msl 30 30 1 16384 TCP MaxSegLifeTime (sec)
cc 0 0 0 1 TTCP function ON/OFF
sack 1 0 0 1 Selective ACK ON/OFF

...and here are the SpeedGuide results:

Code: [Select]
« SpeedGuide.net TCP Analyzer Results »
Tested on: 2019.09.07 17:52
IP address: 50.100.xxx.xxx
 
TCP options string: 020405ac0101040201030302
MSS: 1452
MTU: 1492
TCP Window: 246720 (not multiple of MSS)
RWIN Scaling: 2 bits (2^2=4)
Unscaled RWIN : 61680
Recommended RWINs: 63888, 127776, 255552, 511104, 1022208
BDP limit (200ms): 9869kbps (1234KBytes/s)
BDP limit (500ms): 3948kbps (493KBytes/s)
MTU Discovery: ON
TTL: 48
Timestamps: OFF
SACKs: ON
IP ToS: 00100000 (32)
    Precedence: 001 (priority)
    Delay: 0 (normal delay)
    Throughput: 0 (normal throughput)
    Reliability: 0 (normal reliability)
    Cost: 0 (normal cost)
    Check bit: 0 (correct)
DSCP (DiffServ): CS1 001000 (8) - class 1 (RFC 2474). Similar forwarding behavior to the ToS Precedence field.

Things are basically maxed-out, the OS2 tcpip stack has been working fine (no traps) and that allows my machine to fly when I do have a connection to a really big pipe. Rarely ever do I actually utilize more than 25% of my 1Gig fibre link though.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: David McKenna on September 08, 2019, 04:34:35 pm
 Just tried the TCP Analyzer at Speedguide.net and got a strange result. I have 2 computers running OS/2 (one eCS 2.2 with Intel NIC and one ArcaOS 5.0.4 with Realtek NIC) both attached to the same router and both have identical INETCFG.INI files. I set the RWin value on both to 128480 (MTU 1500 and MSS 1460 on both) which is the highest it can be set on our stack and be a multiple of the MSS. On the eCS computer, when I run the TCP Analyzer it shows that the Default TCP Recieve Window (RWIN) = 128480 as it should. But on the ArcaOS computer the test says it is 65534! If I change RWIN to 64240 on the ArcaOS computer, then the TCP Analyzer says it is 64240 as set. So it seems that somehow the ArcaOS computer RWIN is being throttled to 65534 when set to a higher value? Anyone have any idea what might cause this? Software, hardware? Not sure where to look....
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: ivan on September 08, 2019, 05:33:21 pm
Which NIC driver are you using on both machines?

I assume the ArcaOS one is using the MMRE driver and I have found it problematic in reporting  throughput - I had to stop using IPSPEED.exe because it never reported any speed or throughput, I had no problems while I was using NVETH driver by Nickk.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: David McKenna on September 08, 2019, 05:44:34 pm
 Both are using the latest 32 bit MultiMac from Arca Noae. The eCS machine has also been updated with all of the current ArcaNoae drivers... I'll check the NVETH driver, thanks. <EDIT> looks like NVETH is for NVidia, not Realtek....
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on September 08, 2019, 07:06:05 pm
Are the network cards Realtek? If so, Realtek is known for frequently changing its hardware, often without changing its ID and also they're kind of like WinModems, the driver does a lot of the processing that is done by hardware on other cards.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: David McKenna on September 08, 2019, 09:21:19 pm
Are the network cards Realtek?

 Only the one I am having trouble with. The other is Intel. I think I have a plugin Intel PCIe card laying around somewhere... maybe I'll give it a try...
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on September 08, 2019, 09:39:49 pm
Hi David,

Both are using the latest 32 bit MultiMac from Arca Noae...

I have ran into TRAPs with the newer MMRE drivers, subsequently I went back to the last stable RealTek drivers I had here and all is fine.

PCI reports the following device ID info for my NIC:

Code: [Select]
Vendor 10ECh Realtek Semiconductor Co., Ltd.
 Device 8168h RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller

...is it similar to what you are running? Maybe worth a try to go back to the old RealTek drivers?
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: David McKenna on September 08, 2019, 11:56:07 pm
 Can you specify 'the last stable Realtek driver'?
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on September 09, 2019, 12:06:05 am
Can you specify 'the last stable Realtek driver'?

Ahh....of course, sorry...rushed through that response having focused on the PCI Device_ID, should have posted the driver info to start off with.

OK, here it is:

Code: [Select]
[G:\ibmcom\macs]dir R8169.OS2

The volume label in drive G is OS2.
The Volume Serial Number is 45B6:75C3.
Directory of G:\ibmcom\macs

 9-03-15   3:26p     53724           0  r8169.os2
        1 file(s)      53724 bytes used
                 155185074 K bytes free

[G:\ibmcom\macs]bldlevel r8169.os2
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature:       @#Arca Noae LLC:1.0.2#@##1## 3 Sep 2015 15:26:18      DAZAR1
 ::::2::SVN48@@Realtek 8169/1Gb PCIe MAC Driver Copyright (c) 2014 Arca Noae LLC

Vendor:          Arca Noae LLC
Revision:        1.00.2
Date/Time:       3 Sep 2015 15:26:18
Build Machine:   DAZAR1
FixPak Version:  SVN48
File Version:    1.0.2
Description:     Realtek 8169/1Gb PCIe MAC Driver Copyright (c) 2014 Arca Noae L
LC
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Doug Bissett on September 09, 2019, 08:30:27 am
Quote
r8169.os2

Personally, I wouldn't call that driver "stable", in any sense of the word. I have 6 Realtek gigabit NICS, only 3 of them actually function with that driver. All of them function with MMRE.OS2. If you have a device that doesn't work reliably with MMRE, I would suggest opening an incident, and get MMRE fixed. r8169.os2 is obsolete.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on September 09, 2019, 02:41:01 pm
Hey Doug,

Quote
r8169.os2

Personally, I wouldn't call that driver "stable", in any sense of the word...

Fair, but in my experience that is THE driver that works on my hardware. I find the device driver support rather lacking (even through AN) and given my previous experience of working several tickets (SNAP and Panorma, none of which went anywhere) I decided to avoid that approach altogether.

Look, given that David isn't having much luck with the latest and greatest MMRE drivers it would seem to me like an alternative approach is what's required, even if that means having an obsolete driver that actually works better than the generic & latest MMRE one!  :P
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: David McKenna on September 21, 2019, 03:32:03 pm
 Finally had a chance to try this out again. I installed the R8169.OS2 driver and it worked, but still showed a max RWIN of 65536 at speedguide.net. Also replaced the Realtek NIC with an Intel PCIe Gigabit NIC, but it too showed a max RWIN of 65536. So it seems there is some software at work throttling this? Or could it be wiring? Can the motherboard control this? Next I need to swap around the cables on the switch....
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dave Yeo on September 21, 2019, 06:21:41 pm
Guessing, could be wiring, could be your switch or even upstream.
Here I'm using LTE to connect and my numbers are really limited, RWIN seems to be stuck at 45056, MSS 1370 (changed to 536 after I changed MTU) and weirdest is MTU, 1410 usually but I just did a "ifconfig lan0 mtu 576" and now its stuck at 576 no matter if I try setting it to 1500, 1410, 1200, 600. Didn't try lower.
Here it is likely a limit of using LTE but could be the hub, which I'm cabled to. Hmm, trying on my phone (WiFi) gives the same default numbers (before I played with MTU on the desktop) as I had. Also mentions that Chrome can change the numbers independently of the OS on sockets it creates on some OSes.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on September 24, 2019, 12:59:39 pm
Hi,

Using a new gigabit rj45 cable, my interface only show 100Mbps while it is Gigabit !
Tried MMRE as well as R8189

Some interesting test with nperf compared to windows:
under windows; nperf gives a flat stable 192Mbps   (My provider limiting it to 200Mbps)
under ArcaOs (MMRE or R8189), the nperfs gives me a flutuation between 135Mbps and 45Mbps and never could get the same flat stable result as I have under windows. Any suggestion ? Is it a tcp/ip or driver buffer limitation ?
Cheers
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Doug Bissett on September 24, 2019, 04:54:32 pm
Quote
Using a new gigabit rj45 cable, my interface only show 100Mbps while it is Gigabit !

How are you measuring that? There is a long standing bug (probably a "feature", by now) in NETSTAT that shows gigabit speed as being 100 Mbs, when it is really 1000 Mbs.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Remy on September 24, 2019, 11:38:43 pm
Quote
Using a new gigabit rj45 cable, my interface only show 100Mbps while it is Gigabit !

How are you measuring that? There is a long standing bug (probably a "feature", by now) in NETSTAT that shows gigabit speed as being 100 Mbs, when it is really 1000 Mbs.


netstat -n

This may be possible (thanks for the info) but the problem about throuput going from 45 to 135mbps is not normal. Under windows, I have a very stable 192mbps throuput during the test time (nperf) each time. (with or without IJFW, the fluctuation is the same and changing the driver, not working better.... tcp/ip ?
 
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Doug Bissett on September 25, 2019, 03:17:48 am
Quote
netstat -n

Yeah, that is the one that lies to you.

Quote
but the problem about throuput going from 45 to 135mbps is not normal.

Well, actually it is. That can be caused by a lot of things. The usual cause is bandwidth shaping, by internet suppliers (not necessarily your own).

Quote
Under windows, I have a very stable 192mbps throuput during the test time (nperf) each time. (with or without IJFW, the fluctuation is the same and changing the driver, not working better.... tcp/ip ?

As I said, there are a lot of reasons for the speed to fluctuate. You can do a test between two systems, on your own network, to see what happens (eliminating outside influences). There is a program, called netio, at Hobbes: http://hobbes.nmsu.edu/download/pub/os2/util/network/netio132.zip (http://hobbes.nmsu.edu/download/pub/os2/util/network/netio132.zip). It will work between two computers, and it has support for more than one OS. If that shows the same thing, you need to look at your own setup. If it is good, you need to look outside of your own systems.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on January 12, 2020, 09:58:02 am
My current NBN/HFC connection (Australian) is officially 50/20Mbps but is more like a 38ms 15/2Mbps connection with cached burst max of 32Mbps download and cached sustained download of 17Mbps with maximum non cached download speeds of 3.6Mbps. After much optimisation individually using Windows 7, Zorin OS Ultimate (Ubuntu variant) and OS/2 (without InJoy Firewall) connected directly to the router, I have results that are only relevant to my connection.

My optimised inetcfg.ini file has a maximum Receive, and Send window size of 63360 with an MTU of 1480, and an MSS of 1440. The window size is locked to a maximum of 65534 external to my network no matter which provider I use so can only assume this is a HFC 'thing' so I'm moved to a cheaper ISP with the same results. HFC is the only option available in our area.
 

============== inetcfg.ini ==============

#Inetcfg:   CURRENT   DEFAULT   MINIMUM   MAXIMUM

keepalive   30   7800   0   7800   KeepAlive (sec)
tcpswinsize   63360   32768   1024   246723   TCP SendWindow Size
tcprwinsize   63360   32768   1024   246723   TCP RecvWindow Size
udpswinsize   63360   9216   1024   246723   UDP SendWindow Size
udprwinsize   63360   41600   1024   246723   UDP RecvWindow Size
lingertime   60   120   0   65535   Linger Time (sec)
probecnt   8   8   1   8   Number Of KeepAlive Probes
ipforward   1   0   0   1   IP Forwarding flag ON/OFF
mtudiscover   1   1   0   1   TCP Path MTU Discovery ON/OFF
arpkillc   1200   1200   180   15300   ARP Complete TimeOut (sec)
arpkilli   180   180   60   1200   ARP InComplete TimeOut (sec)
lipcstswinsize   63360   10240   1024   246723   LIPC Stream SendWindow Size
lipcstrwinsize   63360   10240   1024   246723   LIPC Stream RecvWindow Size
lipcdgswinsize   16384   2048   1024   246723   LIPC Dgram SendWindow Size
lipcdgrwinsize   16384   4096   1024   246723   LIPC Dgram RecvWindow Size
synattack   1   0   0   1   SYN Attack Flag ON/OFF
syncookie   0   0   0   1   SYN Cookie Flag ON/OFF
firewall   0   0   0   1   FIREWALL ON/OFF
multidefrt   0   1   0   1   Multiple Default Routes ON/OFF
winscale   1   1   0   1   TCP Window Scale ON/OFF
timestmp   0   0   0   1   TCP TimeStamps ON/OFF
realslow   20   20   1   100   TCP TW Q Slow Timeout ticks
perfhttp   1   1   0   1   Fast Path HTTP ON/OFF
tcpttl      64   64   1   255   TCP  TTL (hops)
udpttl      64   64   1   255   UDP  TTL (hops)
icmpttl      64   64   1   255   ICMP TTL (hops)
fragttl      60   60   1   255   Fragment TTL (sec)
reusetw      1   1   0   1   Reuse Time Wait ON/OFF
msl      30   30   1   16384   TCP MaxSegLifeTime (sec)
cc      0   0   0   1   TTCP function ON/OFF
sack      0   0   0   1   Selective ACK ON/OFF

============== inetcfg.ini ==============

I think the above shows you that the settings you require are very much individually dependent on your personal connection, and the tech used. The best results you can expect are dependent on putting the effort into changing inetcfg.ini settings then updating your system by reloading inetcfg.ini with 'inetcfg -s all' OR using a utility under Windows or Linux to simplify getting the data you need.
RecvWindow size needs to be a multiple of your MSS.

My former ADSL/2+ AnnexM was a 9ms 22/3Mbps PPPoA connection with an MTU of 1500 and a window size around half the maximum of 246723.
Note, all ms results are to the first external hop only, I used a ISP provided router, OR a TPLink AC1200, OR a Cisco SG300-20 Managed switch on each test run to remove anything that may have been attributed to my own hardware.

[edited misspelling, and expanded the last sentence]
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on January 13, 2020, 01:33:11 pm
hi Ian,

on a 30Megabit/sec download / 10Megabit/sec upload full granted (with 8 internet static IPs)
router is a Mikrotik Hex lite, mtu is 1500
how should i calculate the inetcfg values?

at the moment i can get 1,8MegaBytes/sec of download, but if i connect a W10 pc (directly to the Mikrotik router) i get more fast download datarates e.g. 2,5 or more MegaBytes/sec

at the moment on the eCS machine values are theese ones:

config.sys

DEVICE=C:\MPTN\PROTOCOL\SOCKETSK.SYS /GDT:256 /MEM:4096                         
DEVICE=C:\MPTN\PROTOCOL\AFOS2.SYS                                               
DEVICE=C:\MPTN\PROTOCOL\AFINETK.SYS                                                     
RUN=C:\MPTN\BIN\CNTRL.EXE /SM 1024 /LM 720                                                                                                                     

(i've read on EDM/2 that values e.g. buffers can be raised to higher levels, this could help?)

inetcfg:


keepalive 60
tcpswinsize 246723
tcprwinsize 246723
udpswinsize 246723
udprwinsize 246723
lingertime 60
ipforward 1
mtudiscover 1


drivers: eCS PC -> router is an old Intel pro 100 E 100Megabit full duplex (E100B.OS2)

i also suspect that this driver is very old and it's no good for newer connections


thanks

massimo s.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Dariusz Piatkowski on January 14, 2020, 02:47:54 am
Guys,

As I previously posted, I've been running with the SACK flag set to '1' for ever...as in, I can't remember when I didn't actually have that setting in place.

Alright, so now that I've put on some serious time on the new Gigabit setup I figured I'd go back and tweak a few more settings. Specifically, I moved SACK=0 and am looking to see what impact, if any, that has on things like FF.

Now, for what it's worth, my ISP has a pretty fast speedtest box setup, so if anything while not exactly realistic from an overall Internet speed perspective, it does allow me to benchmark my OS/2 setups vs the other machines which are Win boxes that I run at home.

Code: [Select]
Selected Test Server :  20916

Retrieving speedtest.net configuration...
Testing from Managed Network Systems (216.8.142.146)...
Retrieving speedtest.net server list...
Retrieving information for the selected server...
Hosted by MNSi Telecom (Windsor, ON) [7.46 km]: 25.833 ms
Testing download speed..........................................................
......................
Download: 809.64 Mbit/s
Testing upload speed............................................................
....................................
Upload: 588.07 Mbit/s

I am pretty happy. The changes I mentioned above did NOT impact the results, I've been getting these pretty consistently since I moved away from BELL and went to MNSI.NET (local Windsor, ON provider).

Here is what my current inetcfg.ini looks like:

Code: [Select]
#Inetcfg:       CURRENT DEFAULT MINIMUM MAXIMUM

keepalive       60      7800    0       7800    KeepAlive (sec)
tcpswinsize     127776  32768   1024    246723  TCP SendWindow Size
tcprwinsize     127776  32768   1024    246723  TCP RecvWindow Size
udpswinsize     127776  9216    1024    246723  UDP SendWindow Size
udprwinsize     127776  41600   1024    246723  UDP RecvWindow Size
lingertime      120     120     0       65535   Linger Time (sec)
probecnt        8       8       1       8       Number Of KeepAlive Probes
ipforward       0       0       0       1       IP Forwarding flag ON/OFF
mtudiscover     1       1       0       1       TCP Path MTU Discovery ON/OFF
arpkillc        1200    1200    180     15300   ARP Complete TimeOut (sec)
arpkilli        180     180     60      1200    ARP InComplete TimeOut (sec)
lipcstswinsize  10240   10240   1024    246723  LIPC Stream SendWindow Size
lipcstrwinsize  10240   10240   1024    246723  LIPC Stream RecvWindow Size
lipcdgswinsize  2048    2048    1024    246723  LIPC Dgram SendWindow Size
lipcdgrwinsize  4096    4096    1024    246723  LIPC Dgram RecvWindow Size
synattack       0       0       0       1       SYN Attack Flag ON/OFF
syncookie       0       0       0       1       SYN Cookie Flag ON/OFF
firewall        0       0       0       1       FIREWALL ON/OFF
multidefrt      1       1       0       1       Multiple Default Routes ON/OFF
winscale        1       1       0       1       TCP Window Scale ON/OFF
timestmp        1       0       0       1       TCP TimeStamps ON/OFF
realslow        20      20      1       100     TCP TW Q Slow Timeout ticks
perfhttp        1       1       0       1       Fast Path HTTP ON/OFF
tcpttl          64      64      1       255     TCP  TTL (hops)
udpttl          64      64      1       255     UDP  TTL (hops)
icmpttl         64      64      1       255     ICMP TTL (hops)
fragttl         60      60      1       255     Fragment TTL (sec)
reusetw         1       1       0       1       Reuse Time Wait ON/OFF
msl             30      30      1       16384   TCP MaxSegLifeTime (sec)
cc              0       0       0       1       TTCP function ON/OFF
sack            0       0       0       1       Selective ACK ON/OFF
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on January 14, 2020, 04:48:25 am
Hi Massimo,

Quote
on a 30Megabit/sec download / 10Megabit/sec upload full granted (with 8 internet static IPs)
router is a Mikrotik Hex lite, mtu is 1500
how should i calculate the inetcfg values?

MSS (maximum segment size) and MTU (maximum transmission unit)
Your MSS will depend on your packet overhead, ie PPPoE is normally 8 bytes (6 bytes PPPoE + 2 bytes PPP).
my MSS = MTU minus 40(IP header(20 bytes) + TCP header(20 bytes))
=> my MSS = 1440 = 1480 - 40

Receive Window size = multiple of MSS
For an MTU = 1500, assuming your MSS = 1460

Your optimum Receive Window size's will be MSS * 44, with each successive value doubled until you reach your limit.

1460 * 44 = 64240
1460 * 44 * 2 = 128480
1460 * 44 * 2^2 (2 squared) = 256960

As OS/2's maximum Receive Windows size is 246723, you would use the value of 128480
BUT check your MSS, this should be shown somewhere in your router.


Dariusz, SACK flag set to 1 should be fine and is probably preferred on a non server machine that is regularly restarted. The problem's come to the fore when the OS/2 box has a long uptime, hence the recommendation to set SACK to 0 if using OS/2 as a server. I do not think much has checked for the OS/2 stack so assume this will still cause a slow memory leak with the latest stack. It would be nice if I was wrong but I still haven't changed my settings to check, very low on my list of priorities!
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Andi B. on January 14, 2020, 09:45:37 am
Quote
...OR using a utility under Windows or Linux to simplify getting the data you need...
Can you recommend a useful utility for that? I'm not very experienced with Linux but I've an Ubuntu installation on one of my systems which I can try.

I've some WinXP/7/10 notebooks too. Usually I avoid connecting them directly to my ADSL router. But if there is a good Win tool which gives me the proper values for my connection I'll use Win for this task. Thanks.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on January 14, 2020, 06:40:05 pm
Hi Andy,

I used Wireshark.

I'm having a look at whats available via a link way up above at https://www.speedguide.net/analyzer.php (https://www.speedguide.net/analyzer.php)
The links given under "Optimize Your Connection" seem to be quite good.

https://www.speedguide.net/articles/the-tcp-window-latency-and-the-bandwidth-delay-2678 (https://www.speedguide.net/articles/the-tcp-window-latency-and-the-bandwidth-delay-2678) is a good summary.

I will also throw in https://www.ibm.com/support/knowledgecenter/ssw_aix_72/performance/tcp_workload_tuning.html (https://www.ibm.com/support/knowledgecenter/ssw_aix_72/performance/tcp_workload_tuning.html) for more information though much of it is not relevant to OS/2.

There is a Windows based optimiser available here: https://www.speedguide.net/downloads.php (https://www.speedguide.net/downloads.php)
I have not tried this one but the rest of the information they give seems good so worth a try. I originally tried Wireshark under Windows 10, gave up and used Windows 7 were I got some good results.
Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: massimo s. on January 17, 2020, 07:00:34 pm
Ian, i'm sorry

so if i've a 100Megabit/sec FTTC i should set as RWINsize

1022208 (30-100 Mbit lines depending on latency. MSS * 44 * 2^4)

that of course with eCS/AOS is not possible since we have a max value is 246723
so you are saying that we can't have download faster than a 15Megabit/sec connection (i'm just guessing)

but it's not, since here i've 2 servers and on one i get downloads of about 6,9-7MegaBytes/sec

i'm now very very confused :)


Title: Re: Moved to UltraFibre 120 Mb/s but OS/2 chugs at 10 Mb/s...???
Post by: Ian Manners on January 19, 2020, 06:33:37 pm
Hi Massimo.

goto https://www.speedguide.net/analyzer.php (https://www.speedguide.net/analyzer.php)

What does it say your "RWIN Scaling (RFC1323) = 2 bits (scale factor: 2^2=4)" is?

and what value does it give for "Unscaled TCP Receive Window"?

What you are looking for is a MULTIPLE that fits, the size of RWIN is not overly important for us. Ignore the bandwidth * delay product for now. If you are interested you can find more information on that by searching for  RFC1323 and having a read.