Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!msunews!agate!ihnp4.ucsd.edu!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!news.maxwell.syr.edu!news.apfel.de!fu-berlin.de!news.belwue.de!LF.net!pi From: pi@complx.LF.net (Kurt Jaeger) Newsgroups: comp.unix.bsd.freebsd.misc Subject: 2.1.7 && adaptec 2940U[W] crashing Date: 20 Mar 1997 21:19:16 GMT Organization: LF.net GmbH Lines: 54 Message-ID: <5gs9kk$5bu$1@news.LF.net> NNTP-Posting-Host: complx.lf.net Xref: euryale.cc.adfa.oz.au comp.unix.bsd.freebsd.misc:37499 Hi! We recently upgraded our news server to 2.1.7 (it was 2.1.0 before that), because with 2.1.0 it consistently hung (no panic). That problem was solved with 2.1.7, it now crashes and reboots reliably after approx. 24 hours of operations. Error messages are something along the following: [crash 1] Mar 13 18:59:39 news /kernel: sd0(ahc0:0:0): timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0x0 Mar 13 18:59:39 news /kernel: SEQADDR == 0x11 [crash 2] Mar 13 23:35:53 news /kernel: sd1(ahc0:1:0): timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0x0 Mar 13 23:35:53 news /kernel: SEQADDR == 0xd Mar 13 23:41:09 news /kernel: panic: ahc0: Timed-out command times out again At least its not a specific disk that causes that panic. All the SCSI features (disconnect/reconnect, scsi speed > 5 MB/sec etc) were disabled after that, no effect. There are 4 disks in the system. We switched controllers (2940 first, 2940UW after), no difference. I compared the aic7xxx.c code of the 2.1.0 kernel with the 2.1.7 one. Big difference -- maybe the reason the older kernel hung was because it was not able to handle additional timeouts in a reasonable way ? The new one at least tries it. Anyway, the next crash didn't take long, now yet another drive, and it looks like it was pretty confusing to the buffer cache. --------- sd2(ahc0:2:0): timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0x0 SEQADDR == 0x8 pid 25152 (in.nnrpd), uid 23: exited on sig\M^?\^Ganic: ahc0: Timed-out command times out again syncing disks... 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 78 giv ing up Automatic reboot in 15 seconds - press a key on the console to abort --------- The next thing I'd like to do is a) shorten the scsi-cable (currently approx. 1.5 m) and b) to enable savecore -- anyone with a guide on how to use its result 8-} ? -- MfG/Best regards, Kurt Jaeger 23 years to go ! LF.net GmbH pi@LF.net Vor dem Lauch 23 fon +49 711 90074-23 D-70567 Stuttgart fax +49 711 7289041