Date: Tue, 18 Jun 1996 09:59:52 -0400 (EDT) From: Rashid Karimov <rashid@rk.ios.com> To: freebsd-stable@freebsd.org Cc: questions@freebsd.org Subject: AHC/SCSI stability (resets/timeouts on SCSI) Message-ID: <199606181359.JAA13394@rk.ios.com>
next in thread | raw e-mail | index | archive | help
Hi there folx, I've changed the topic of some other thread here ( famous trap 12 one) to something more relevant to the Q I have. The cites below are from Justin's answer on kernel config issues. I run rather large CCD PC here - it has 3 AHA2940W and 6 HDDs in it ( it's a news server) . Works very fast - reads/writes at ~18MBs. It can ran up to a few weeks w/o any problems, but every once in a while I would see these messages in the messages file: Jun 11 21:49:45 news /kernel: ahc2: Issued Channel A Bus Reset #3. 0 SCBs aborted Jun 11 21:49:45 news /kernel: ahc2: target 1, lun 0 (sd5) timed out Jun 11 21:49:45 news /kernel: ahc2: Issued Channel A Bus Reset #3. 0 SCBs aborted /kernel: sd5(ahc2:1:0): UNIT ATTENTION asc:29,0 un 11 21:49:49 news /kernel: sd5(ahc2:1:0): Power on, reset, or bus device reset occurred field replaceable uni t: 1 Jun 11 21:49:49 news /kernel: , retries:4 Jun 11 21:49:49 news /kernel: biodone: buffer already done and machine would reboot or hang :( Sometime all three ahc* would complain ... The ahc driver is from 2.1-release I think. Is there anything I can do to make the machine more stable ? I want to give a try to ahc driver from latest 2.2.-SNAP to see how it goes . >Here is my kernel configuration: > > >controller ahc0 > >controller ahc1 > > You only need one of these. One ahc entry will attach to as many cards > as it can find in your system. I have 3 entries actually - can it somehow affect stability of the SCSI system ? I can't go with whole new kernel though - the ccd is now made as LKM and is not backward compatible as I heard ? Is there a chance I can use old ccd code with 2.2-0612-SNAP ? > > >Basically I stripped everything I didn't need in the hopes > >of expurgating the problem. > > Not quite everything. > > >An issue with the 2940W? I did have to disable wide > >transfers on the wide drive and sync negotiation for both (not sure > >which fixed it) to make the narrow drive stop rebooting the SCSI bus > >on every drive access, but that was a dozen sups and kernels ago, > >haven't had time to play with it since. > > Please confirm that you no longer need to disable wide negotiation or > lower the sync rate with the latest driver. This should be fixed. > > >Later, > >Jeff > > -- > Justin T. Gibbs Rashid.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199606181359.JAA13394>