Re: ACPI??? was - Re: -current TCP performance hosed?
On Mon, Sep 17, 2001 at 02:55:16PM +0200, Geoff Rehmet wrote: > On Mon, Sep 17, 2001 at 11:55:48AM +0200, Oliver Fromme wrote: > > Well, duplex mismatch can result in asymmetric behaviour. > > > > It could be a problem with the transmit part of the driver. > > You didn't mention what kind of NIC and driver you use, > > IIRC. Maybe trying a different NIC could indicate whether > > this is a driver problem or something else. > I intend to change my NIC. That will just take a little more time to arrange. I've found the culprit - after moving my main network interface from ep0 to xl0, things suddenly work a lot better. Thus, it is probably either the card or the driver. (Admittedly, the card is quite ancient.) Geoff. -- Geoff Rehmet, Internet Solutions tel: +27-11-283-5462, fax: +27-11-283-5401 mobile: +27-83-292-5800 email: [EMAIL PROTECTED] URL: http://www.is.co.za To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
On Mon, Sep 17, 2001 at 11:55:48AM +0200, Oliver Fromme wrote: > Well, duplex mismatch can result in asymmetric behaviour. > > It could be a problem with the transmit part of the driver. > You didn't mention what kind of NIC and driver you use, > IIRC. Maybe trying a different NIC could indicate whether > this is a driver problem or something else. I intend to change my NIC. That will just take a little more time to arrange. Geoff. -- Geoff Rehmet, Internet Solutions tel: +27-11-283-5462, fax: +27-11-283-5401 mobile: +27-83-292-5800 email: [EMAIL PROTECTED] URL: http://www.is.co.za To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
Geoff Rehmet <[EMAIL PROTECTED]> wrote: > On Sun, Sep 16, 2001 at 12:37:53PM +0200, Oliver Fromme wrote: > > Have you verified that the duplex setting of your network > > interface is correct? It should be set to half-duplex if > > the machine is connected to a hub. Don't trust autoselect. > > Check the collision LEDs at the card (if present) and at > > the hub during data transfer. If everything looks OK, try > > putting a different card into that machine. > > My ethernet card is definitely running half duplex. Also, > as I mentioned, as a client, the box behaves fine, but not > as a server. Well, duplex mismatch can result in asymmetric behaviour. It could be a problem with the transmit part of the driver. You didn't mention what kind of NIC and driver you use, IIRC. Maybe trying a different NIC could indicate whether this is a driver problem or something else. Regards Oliver -- Oliver Fromme, secnetix GmbH & Co KG, Oettingenstr. 2, 80538 München Any opinions expressed in this message may be personal to the author and may not necessarily reflect the opinions of secnetix in any way. "All that we see or seem is just a dream within a dream" (E. A. Poe) To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
On Sun, Sep 16, 2001 at 12:37:53PM +0200, Oliver Fromme wrote: > > Have you verified that the duplex setting of your network > interface is correct? It should be set to half-duplex if > the machine is connected to a hub. Don't trust autoselect. > Check the collision LEDs at the card (if present) and at > the hub during data transfer. If everything looks OK, try > putting a different card into that machine. My ethernet card is definitely running half duplex. Also, as I mentioned, as a client, the box behaves fine, but not as a server. Geoff. -- Geoff Rehmet, Internet Solutions tel: +27-11-283-5462, fax: +27-11-283-5401 mobile: +27-83-292-5800 email: [EMAIL PROTECTED] URL: http://www.is.co.za To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
Geoff Rehmet <[EMAIL PROTECTED]> wrote: > No, I can't say for certain when this started. In fact, reverting to a > kernel from June 27 still shows the same problem. > > However, I have done the following exercise, with three machines, > two of which sit on our internal LAN together, on the same hub, with > the third sitting on our public network (in our hosting room). > [...] > At this point, this seems, from the empirical evidence, to have nothing > to do with ACPI. This is probably a dumb question, but just to make sure ... Have you verified that the duplex setting of your network interface is correct? It should be set to half-duplex if the machine is connected to a hub. Don't trust autoselect. Check the collision LEDs at the card (if present) and at the hub during data transfer. If everything looks OK, try putting a different card into that machine. I'm running -current with some DEC clone NIC connected to a FastEthernet switch (running full-duplex), and there's no TCP performance problem. Regards Oliver -- Oliver Fromme, secnetix GmbH & Co KG, Oettingenstr. 2, 80538 München Any opinions expressed in this message may be personal to the author and may not necessarily reflect the opinions of secnetix in any way. "All that we see or seem is just a dream within a dream" (E. A. Poe) To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
On Sat, Sep 15, 2001 at 09:43:57AM -0400, Bosko Milekic wrote: > > On Sat, Sep 15, 2001 at 01:08:28AM +0200, Geoff Rehmet wrote: > > You said this happens with a few-day-old kernel. Is this an > implication that it didn't happen before? Can you say, with a > certain level of certainty exactly when this started happening, > especially if it's something that only started to occur very > recently, this may be a worthy piece of information. Thanks. No, I can't say for certain when this started. In fact, reverting to a kernel from June 27 still shows the same problem. However, I have done the following exercise, with three machines, two of which sit on our internal LAN together, on the same hub, with the third sitting on our public network (in our hosting room). Configs being: hangdog:- 5.0-CURRENT, 128M RAM, PII-266 (internal LAN) deadpoint: - 4.3-STABLE, 128M RAM, PI-166 (internal LAN) illuminati: - 4.2-RELEASE, 128M RAM, PII-333 (public network) hangdog and deadpoint sit on the same hub at my desk. There are 3 router hops from my hub to the hosting network - one of those hops being a firewall. All three machines are running IPFW. I did a test of downloading a 240M odd file from one machine to the other as a test. Everything looked fine, until I used hangdog (my -CURRENT box) as the server. I aborted that download, as it was too painfully slow. The transfer rates were as follows: ServerClient ---- illuminati -> deadpoint (4.2 -> 4.3)852kBps illuminati -> hangdog (4.2 -> 5.0)788kBps hangdog-> deadpoint (5.0 -> 4.3) 83kBps Thus, it appears that -CURRENT is still doing OK as a client, but is struggling badly as a server. At this point, this seems, from the empirical evidence, to have nothing to do with ACPI. Geoff. == deadpoint:/usr/tmp# fetch http://illuminati/~geoff/V379243.GHO Receiving V379243.GHO (255742011 bytes): 100% 255742011 bytes transferred in 292.8 seconds (852.90 kBps) deadpoint:/usr/tmp# rm V379243.GHO deadpoint:/usr/tmp# uname -a FreeBSD deadpoint.is.co.za 4.3-STABLE FreeBSD 4.3-STABLE #2: Thu Jul 19 14:15:37 SAST 2001 [EMAIL PROTECTED]:/usr/src/sys/compile/DEADPOINT i386 deadpoint:/usr/tmp# deadpoint:/usr/tmp# dmesg | more Copyright (c) 1992-2001 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 4.3-STABLE #2: Thu Jul 19 14:15:37 SAST 2001 [EMAIL PROTECTED]:/usr/src/sys/compile/DEADPOINT Timecounter "i8254" frequency 1193182 Hz Timecounter "TSC" frequency 165790172 Hz CPU: Pentium/P55C (165.79-MHz 586-class CPU) === hangdog:~% fetch http://illuminati/~geoff/V379243.GHO Receiving V379243.GHO (255742011 bytes): 100% 255742011 bytes transferred in 316.8 seconds (788.40 kBps) hangdog:~% hangdog:~% dmesg | more Copyright (c) 1992-2001 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 5.0-CURRENT #0: Tue Sep 11 14:15:17 SAST 2001 [EMAIL PROTECTED]:/usr/src/sys/i386/compile/HANGDOG Timecounter "i8254" frequency 1193182 Hz Timecounter "TSC" frequency 267274474 Hz CPU: Pentium II/Pentium II Xeon/Celeron (267.27-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x634 Stepping = 4 === deadpoint:/usr/tmp# fetch http://hangdog/V379243.GHO Receiving V379243.GHO (255742011 bytes): 4%^C 10948608 bytes transferred in 128.6 seconds (83.11 kBps) fetch: transfer interrupted deadpoint:/usr/tmp# = illuminati:~% dmesg | more Copyright (c) 1992-2000 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 4.2-RELEASE #2: Mon Feb 26 14:24:51 SAST 2001 [EMAIL PROTECTED]:/usr/src/sys/compile/CUSTOM Timecounter "i8254" frequency 1193182 Hz Timecounter "TSC" frequency 332388500 Hz CPU: Pentium II/Pentium II Xeon/Celeron (332.39-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x650 Stepping = 0 -- Geoff Rehmet, Internet Solutions tel: +27-11-283-5462, fax: +27-11-283-5401 mobile: +27-83-292-5800 email: [EMAIL PROTECTED] URL: http://www.is.co.za To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
On Sat, Sep 15, 2001 at 01:08:28AM +0200, Geoff Rehmet wrote: > It seems that, on Fri, Sep 14, 2001 at 03:18:47PM -0700, > in message <[EMAIL PROTECTED]>, Mike Smith wrote: > > > Nope, no debug options, but I am getting loads of > > > > > > microuptime() went backwards (29804.3839847 -> 29804.925730) > > > > ALi chipset? Try turning off the ACPI timer if you haven't already; > > > > set debug.acpi.disable="timer" > > > > at the loader prompt. If this works, please let me know (with ACPI in the > > subject line so I don't miss it). > Tried that, also tried "set hint.acpi.0.disable=1" - neither > had any effect. > The kernel is not compiled with apm... You said this happens with a few-day-old kernel. Is this an implication that it didn't happen before? Can you say, with a certain level of certainty exactly when this started happening, especially if it's something that only started to occur very recently, this may be a worthy piece of information. Thanks. > Geoff. > > -- > Geoff Rehmet, Internet Solutions > tel: +27-11-283-5462, fax: +27-11-283-5401 mobile: +27-83-292-5800 > email: [EMAIL PROTECTED], [EMAIL PROTECTED] > URL: http://www.is.co.za > -- Bosko Milekic [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
> FYI -- this has happened on my laptop (Toshiba Tecra 8000) ever since the > ACPI was first introduced (months and months ago) What has happened, specifically? If you disable the timer, as Geoff did, does it still happen? Details, details. > On Fri, 14 Sep 2001, Mike Smith wrote: > > > > It seems that, on Fri, Sep 14, 2001 at 03:18:47PM -0700, > > > in message <[EMAIL PROTECTED]>, Mike Smith wrote: > > > > > Nope, no debug options, but I am getting loads of > > > > > > > > > > microuptime() went backwards (29804.3839847 -> 29804.925730) > > > > > > > > ALi chipset? Try turning off the ACPI timer if you haven't already; > > > > > > > > set debug.acpi.disable="timer" > > > > > > > > at the loader prompt. If this works, please let me know (with ACPI in t > he > > > > subject line so I don't miss it). > > > > > > Tried that, also tried "set hint.acpi.0.disable=1" - neither > > > had any effect. > > > > Can you explicitly disable the ACPI module? > > > > unset acpi_load > > > > at the loader prompt. If it still happens with that, then something else is > > b0rked and I can stop panicking. 8) > > > > To Unsubscribe: send mail to [EMAIL PROTECTED] > > with "unsubscribe freebsd-current" in the body of the message > > > > -- >_ __ ___ ___ ___ ___ > Wesley N Morgan _ __ ___ | _ ) __| \ > [EMAIL PROTECTED] _ __ | _ \._ \ |) | > FreeBSD: The Power To Serve _ |___/___/___/ > 6bone: 3ffe:1ce3:7::b4ff:fe53:c297 > Hi! I'm a .signature virus! Copy me into your ~/.signature to help me spread! > To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
FYI -- this has happened on my laptop (Toshiba Tecra 8000) ever since the ACPI was first introduced (months and months ago) On Fri, 14 Sep 2001, Mike Smith wrote: > > It seems that, on Fri, Sep 14, 2001 at 03:18:47PM -0700, > > in message <[EMAIL PROTECTED]>, Mike Smith wrote: > > > > Nope, no debug options, but I am getting loads of > > > > > > > > microuptime() went backwards (29804.3839847 -> 29804.925730) > > > > > > ALi chipset? Try turning off the ACPI timer if you haven't already; > > > > > > set debug.acpi.disable="timer" > > > > > > at the loader prompt. If this works, please let me know (with ACPI in the > > > subject line so I don't miss it). > > > > Tried that, also tried "set hint.acpi.0.disable=1" - neither > > had any effect. > > Can you explicitly disable the ACPI module? > > unset acpi_load > > at the loader prompt. If it still happens with that, then something else is > b0rked and I can stop panicking. 8) > > To Unsubscribe: send mail to [EMAIL PROTECTED] > with "unsubscribe freebsd-current" in the body of the message > -- _ __ ___ ___ ___ ___ Wesley N Morgan _ __ ___ | _ ) __| \ [EMAIL PROTECTED] _ __ | _ \._ \ |) | FreeBSD: The Power To Serve _ |___/___/___/ 6bone: 3ffe:1ce3:7::b4ff:fe53:c297 Hi! I'm a .signature virus! Copy me into your ~/.signature to help me spread! To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: ACPI??? was - Re: -current TCP performance hosed?
> It seems that, on Fri, Sep 14, 2001 at 03:18:47PM -0700, > in message <[EMAIL PROTECTED]>, Mike Smith wrote: > > > Nope, no debug options, but I am getting loads of > > > > > > microuptime() went backwards (29804.3839847 -> 29804.925730) > > > > ALi chipset? Try turning off the ACPI timer if you haven't already; > > > > set debug.acpi.disable="timer" > > > > at the loader prompt. If this works, please let me know (with ACPI in the > > subject line so I don't miss it). > > Tried that, also tried "set hint.acpi.0.disable=1" - neither > had any effect. Can you explicitly disable the ACPI module? unset acpi_load at the loader prompt. If it still happens with that, then something else is b0rked and I can stop panicking. 8) To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message