Re: bce(4) and rx errors
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, Mike, Mike Jakubik wrote: > On Sun, December 14, 2008 4:36 am, Xin LI wrote: >> -BEGIN PGP SIGNED MESSAGE- >> Hash: SHA1 >> >> Hi, Gents, >> >> I have not yet talked this with David but it looks like this patch would >> make it disappear. >> >> Cheers, > > I have been running my production systems for about a week now with this > patch, interrupts are low and rx errors reports are gone, thanks Xin. > > P.S. I'm still having problems with my application, but at this point i'm > not sure whether it has anything to do with the driver, for what its worth > here is the error which causes it to stop accepting new connection. > > wrapper.log.15:INFO | jvm 1| 2008/12/17 18:49:12 | Exception in > thread "Thread-0" java.nio.channels.IllegalBlockingModeException > wrapper.log.15:INFO | jvm 1| 2008/12/17 18:49:12 | at > sun.nio.ch.ServerSocketAdaptor.accept(ServerSocketAdaptor.java:86) > wrapper.log.15:INFO | jvm 1| 2008/12/17 18:49:12 | at > com.precyse.chat.service.connector.ServerSocketChannelConnector.registerSelection(ServerSocketChannelConnector.java:404) > wrapper.log.15:INFO | jvm 1| 2008/12/17 18:49:12 | at > com.precyse.chat.service.connector.ServerSocketChannelConnector.run(ServerSocketChannelConnector.java:148) Thanks for the feedback. Could you check netstat -m and/or vmstat -z to see if there is any mbuf exhaustion issue? Cheers, - -- Xin LI http://www.delphij.net/ FreeBSD - The Power to Serve! -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (FreeBSD) iEYEARECAAYFAklRHYoACgkQi+vbBBjt66AlTACeOz59K8geIlaJTIjzf6dB0xHR UwMAoLCy9bTTha8fu0Y6v1SzzMk5JuOH =wh5l -END PGP SIGNATURE- ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: bce(4) and rx errors
On Sun, December 14, 2008 4:36 am, Xin LI wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Hi, Gents, > > I have not yet talked this with David but it looks like this patch would > make it disappear. > > Cheers, I have been running my production systems for about a week now with this patch, interrupts are low and rx errors reports are gone, thanks Xin. P.S. I'm still having problems with my application, but at this point i'm not sure whether it has anything to do with the driver, for what its worth here is the error which causes it to stop accepting new connection. wrapper.log.15:INFO | jvm 1| 2008/12/17 18:49:12 | Exception in thread "Thread-0" java.nio.channels.IllegalBlockingModeException wrapper.log.15:INFO | jvm 1| 2008/12/17 18:49:12 |at sun.nio.ch.ServerSocketAdaptor.accept(ServerSocketAdaptor.java:86) wrapper.log.15:INFO | jvm 1| 2008/12/17 18:49:12 |at com.precyse.chat.service.connector.ServerSocketChannelConnector.registerSelection(ServerSocketChannelConnector.java:404) wrapper.log.15:INFO | jvm 1| 2008/12/17 18:49:12 |at com.precyse.chat.service.connector.ServerSocketChannelConnector.run(ServerSocketChannelConnector.java:148) ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: bce(4) and rx errors
On Sun, Dec 14, 2008 at 01:36:35AM -0800, Xin LI wrote: > I have not yet talked this with David but it looks like this patch would > make it disappear. Applied to 7.1-RC1, and rx errors are gone. Thank you! Jeff ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: bce(4) and rx errors
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, Gents, I have not yet talked this with David but it looks like this patch would make it disappear. Cheers, - -- Xin LI http://www.delphij.net/ FreeBSD - The Power to Serve! -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (FreeBSD) iEYEARECAAYFAklE06IACgkQi+vbBBjt66DwjgCeJ1RO+gL3mu13zAzXpPsbDi/R /BwAmQFAij74yusRq+KbcSNb3BMbLzNX =1zx3 -END PGP SIGNATURE- Index: if_bce.c === --- if_bce.c(revision 186076) +++ if_bce.c(working copy) @@ -7408,7 +7408,6 @@ (u_long) sc->stat_IfInMBUFDiscards + (u_long) sc->stat_Dot3StatsAlignmentErrors + (u_long) sc->stat_Dot3StatsFCSErrors + - (u_long) sc->stat_IfInFramesL2FilterDiscards + (u_long) sc->stat_IfInRuleCheckerDiscards + (u_long) sc->stat_IfInFTQDiscards + (u_long) sc->com_no_buffers; ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: bce(4) and rx errors
On 12/10/08, Vlad GALU <[EMAIL PROTECTED]> wrote: > On 12/10/08, Mike Jakubik <[EMAIL PROTECTED]> wrote: > > On Wed, December 10, 2008 11:03 am, Jeff Blank wrote: > > > On Wed, Dec 10, 2008 at 04:59:26PM +0200, Vlad GALU wrote: > > >> I have an application pulling about 220Kpps from a bce(4) card > > >> (details below). At what seems to be random times, errors start > > >> showing up on that interface (I'm watching it with netstat -w1 -I), so > > >> about 10% of the initial 220Kpps is reported as errors. > > > > > > I'm also seeing a pretty steady stream of errors on both bce > > > interfaces in a Dell PowerEdge 2950 III. In my case, the source > > > is RELENG_7_1 from ~14:00 UTC yesterday (9 Dec). Throughput does not > > > seem to be affected. "sysctl -a | egrep -i 'bce.*err'" yields all > > > zeroes, for whatever that's worth. > > > > > > See the "RELENG_7_1: bce driver change generating too much interrupts ?" > > thread. This problem as surfaced since the recent bce driver changes. > > > Thanks Mike, I'll give it a shot. Indeed, the errors seem to have gone away after rolling back the driver, as Xin Li suggested in another thread. Sorry for the noise! -- ~/.signature: no such file or directory ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: bce(4) and rx errors
On 12/10/08, Mike Jakubik <[EMAIL PROTECTED]> wrote: > On Wed, December 10, 2008 11:03 am, Jeff Blank wrote: > > On Wed, Dec 10, 2008 at 04:59:26PM +0200, Vlad GALU wrote: > >> I have an application pulling about 220Kpps from a bce(4) card > >> (details below). At what seems to be random times, errors start > >> showing up on that interface (I'm watching it with netstat -w1 -I), so > >> about 10% of the initial 220Kpps is reported as errors. > > > > I'm also seeing a pretty steady stream of errors on both bce > > interfaces in a Dell PowerEdge 2950 III. In my case, the source > > is RELENG_7_1 from ~14:00 UTC yesterday (9 Dec). Throughput does not > > seem to be affected. "sysctl -a | egrep -i 'bce.*err'" yields all > > zeroes, for whatever that's worth. > > > See the "RELENG_7_1: bce driver change generating too much interrupts ?" > thread. This problem as surfaced since the recent bce driver changes. Thanks Mike, I'll give it a shot. -- ~/.signature: no such file or directory ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: bce(4) and rx errors
On Wed, December 10, 2008 11:03 am, Jeff Blank wrote: > On Wed, Dec 10, 2008 at 04:59:26PM +0200, Vlad GALU wrote: >> I have an application pulling about 220Kpps from a bce(4) card >> (details below). At what seems to be random times, errors start >> showing up on that interface (I'm watching it with netstat -w1 -I), so >> about 10% of the initial 220Kpps is reported as errors. > > I'm also seeing a pretty steady stream of errors on both bce > interfaces in a Dell PowerEdge 2950 III. In my case, the source > is RELENG_7_1 from ~14:00 UTC yesterday (9 Dec). Throughput does not > seem to be affected. "sysctl -a | egrep -i 'bce.*err'" yields all > zeroes, for whatever that's worth. See the "RELENG_7_1: bce driver change generating too much interrupts ?" thread. This problem as surfaced since the recent bce driver changes. ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: bce(4) and rx errors
On Wed, Dec 10, 2008 at 04:59:26PM +0200, Vlad GALU wrote: > I have an application pulling about 220Kpps from a bce(4) card > (details below). At what seems to be random times, errors start > showing up on that interface (I'm watching it with netstat -w1 -I), so > about 10% of the initial 220Kpps is reported as errors. I'm also seeing a pretty steady stream of errors on both bce interfaces in a Dell PowerEdge 2950 III. In my case, the source is RELENG_7_1 from ~14:00 UTC yesterday (9 Dec). Throughput does not seem to be affected. "sysctl -a | egrep -i 'bce.*err'" yields all zeroes, for whatever that's worth. [EMAIL PROTECTED]:0:0:0:class=0x06 card=0x80868086 chip=0x25c08086 rev=0x12 hdr=0x00 vendor = 'Intel Corporation' device = '5000X Chipset Memory Controller Hub' class = bridge subclass = HOST-PCI [EMAIL PROTECTED]:0:2:0:class=0x060400 card=0x chip=0x25e28086 rev=0x12 hdr=0x01 vendor = 'Intel Corporation' device = '5000 Series Chipset PCIe x4 Port 2' class = bridge subclass = PCI-PCI [EMAIL PROTECTED]:0:3:0:class=0x060400 card=0x chip=0x25e38086 rev=0x12 hdr=0x01 vendor = 'Intel Corporation' device = '5000 Series Chipset PCIe x4 Port 3' class = bridge subclass = PCI-PCI [EMAIL PROTECTED]:0:4:0:class=0x060400 card=0x chip=0x25f88086 rev=0x12 hdr=0x01 vendor = 'Intel Corporation' device = '5000 Series Chipset PCIe x8 Port 4-5' class = bridge subclass = PCI-PCI [EMAIL PROTECTED]:0:5:0:class=0x060400 card=0x chip=0x25e58086 rev=0x12 hdr=0x01 vendor = 'Intel Corporation' device = '5000 Series Chipset PCIe x4 Port 5' class = bridge subclass = PCI-PCI [EMAIL PROTECTED]:0:6:0:class=0x060400 card=0x chip=0x25f98086 rev=0x12 hdr=0x01 vendor = 'Intel Corporation' device = '5000 Series Chipset PCIe x8 Port 6-7' class = bridge subclass = PCI-PCI [EMAIL PROTECTED]:0:7:0:class=0x060400 card=0x chip=0x25e78086 rev=0x12 hdr=0x01 vendor = 'Intel Corporation' device = '5000 Series Chipset PCIe x4 Port 7' class = bridge subclass = PCI-PCI [EMAIL PROTECTED]:0:16:0: class=0x06 card=0x01b21028 chip=0x25f08086 rev=0x12 hdr=0x00 vendor = 'Intel Corporation' device = '5000 Series Chipset Error Reporting Registers' class = bridge subclass = HOST-PCI [EMAIL PROTECTED]:0:16:1: class=0x06 card=0x01b21028 chip=0x25f08086 rev=0x12 hdr=0x00 vendor = 'Intel Corporation' device = '5000 Series Chipset Error Reporting Registers' class = bridge subclass = HOST-PCI [EMAIL PROTECTED]:0:16:2: class=0x06 card=0x01b21028 chip=0x25f08086 rev=0x12 hdr=0x00 vendor = 'Intel Corporation' device = '5000 Series Chipset Error Reporting Registers' class = bridge subclass = HOST-PCI [EMAIL PROTECTED]:0:17:0: class=0x06 card=0x80868086 chip=0x25f18086 rev=0x12 hdr=0x00 vendor = 'Intel Corporation' device = '5000 Series Chipset Reserved Registers' class = bridge subclass = HOST-PCI [EMAIL PROTECTED]:0:19:0: class=0x06 card=0x80868086 chip=0x25f38086 rev=0x12 hdr=0x00 vendor = 'Intel Corporation' device = '5000 Series Chipset Reserved Registers' class = bridge subclass = HOST-PCI [EMAIL PROTECTED]:0:21:0: class=0x06 card=0x80868086 chip=0x25f58086 rev=0x12 hdr=0x00 vendor = 'Intel Corporation' device = '5000 Series Chipset FBD Registers' class = bridge subclass = HOST-PCI [EMAIL PROTECTED]:0:22:0: class=0x06 card=0x80868086 chip=0x25f68086 rev=0x12 hdr=0x00 vendor = 'Intel Corporation' device = '5000 Series Chipset FBD Registers' class = bridge subclass = HOST-PCI [EMAIL PROTECTED]:0:28:0: class=0x060400 card=0x01b21028 chip=0x26908086 rev=0x09 hdr=0x01 vendor = 'Intel Corporation' device = '631xESB/632xESB/3100 PCIe Root Port 1' class = bridge subclass = PCI-PCI [EMAIL PROTECTED]:0:29:0: class=0x0c0300 card=0x01b21028 chip=0x26888086 rev=0x09 hdr=0x00 vendor = 'Intel Corporation' device = '631xESB/632xESB/3100 Chipset USB Universal Host Controller' class = serial bus subclass = USB [EMAIL PROTECTED]:0:29:1: class=0x0c0300 card=0x01b21028 chip=0x26898086 rev=0x09 hdr=0x00 vendor = 'Intel Corporation' device = '631xESB/632xESB/3100 Chipset USB Universal Host Controller' class = serial bus subclass = USB [EMAIL PROTECTED]:0:29:2: class=0x0c0300 card=0x01b21028 chip=0x268a8086 rev=0x09 hdr=0x00 vendor = 'Intel Corporation' device = '631xESB/632xESB/3100 Chipset USB Universal Host Controller' class =
bce(4) and rx errors
Hello. Sorry for crossposting, but I wasn't sure which mailing list was the most appropriate for this email. I have an application pulling about 220Kpps from a bce(4) card (details below). At what seems to be random times, errors start showing up on that interface (I'm watching it with netstat -w1 -I), so about 10% of the initial 220Kpps is reported as errors. Bringing the interface down and then back up clears the errors, but they do reappear at a later time. Before they reappear, the systems manages to pull the full 220Kpps as before. This is a temporary setup, we'll very soon use an Intel fiber card, but I thought this issue was worth mentioning, as I don't think it's a hardware problem (the switch also reports no errors). The system is running a fresh (yesterday's) RELENG_7. The card is onboard, on a HP DL380 G5. Here's the pciconf output: -- cut here -- [EMAIL PROTECTED]:2:0:0: class=0x060400 card=0x chip=0x01031166 rev=0xc3 hdr=0x01 vendor = 'ServerWorks (Was: Reliance Computer Corp)' device = 'BCM5715 Broadcom dual gigabit, pci bridge' class = bridge subclass = PCI-PCI [EMAIL PROTECTED]:3:0:0:class=0x02 card=0x7038103c chip=0x164c14e4 rev=0x12 hdr=0x00 vendor = 'Broadcom Corporation' device = '5708C Broadcom NetXtreme II Gigabit Ethernet Adapter' class = network subclass = ethernet -- and here -- Regards, Vlad -- ~/.signature: no such file or directory ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"