Date: Thu, 16 Sep 2004 14:16:50 -0700 From: Julian Elischer <julian@elischer.org> To: Daniel Eriksson <daniel_k_eriksson@telia.com> Cc: 'Ariff Abdullah' <skywizard@MyBSD.org.my> Subject: Re: ad0: TIMEOUT - READ_DMA retrying (2 retries left) LBA=207594611 Message-ID: <414A02C2.4010009@elischer.org> In-Reply-To: <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA0VcX9IoJqUaXPS8MjT1PdsKAAAAQAAAApNeCp9Vb6EGg0rWg4vStwwEAAAAA@telia.com> References: <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA0VcX9IoJqUaXPS8MjT1PdsKAAAAQAAAApNeCp9Vb6EGg0rWg4vStwwEAAAAA@telia.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Daniel Eriksson wrote: >Ariff Abdullah wrote: > > > >>One more thing to consider, the default scheduler. I found that those >>errors occured with SCHED_4BSD (PREEMPTION or NOT), while SCHED_ULE >>(of course without PREEMPTION, or *else*), nothing such that. >> >> > >I am using SCHED_4BSD without PREEMPTION. Given the recent stability >problems with ULE I haven't even toyed with the idea of trying it since the >machine in question is a production machine. > >I'm curious if this problem is in any way related to my other problem: I >cannot use ataraid and gstripe at the same time. If I start gstripe I get an >interrupt storm followed by timeouts that tears down the ataraid arrays. >Next time I feel lucky I'll try gstripe again (Pawel Jakub Dawidek told me >how to get some more debugging output from it). Only reason I haven't done >it yet is because the machine has been very nice and stable for the last two >weeks. > > >Which reminds me: there was an interrupt storm preceeding the messages I >got. I missed copying that line in my original "me-too" message: > >Interrupt storm detected on "irq9: acpi0"; throttling interrupt source >ad0: TIMEOUT - READ_DMA retrying (2 retries left) LBA=75822564 >ad0: WARNING - READ_DMA no interrupt but good status >... > I have a suspicion that some of my attempts to stabilse preemption in 4bsd may have a hand in this.. I'm looking at some traces now that may be relevent.. you might try ULE now too as some fixes went in that may have influenced that. > >/Daniel Eriksson > > >_______________________________________________ >freebsd-current@freebsd.org mailing list >http://lists.freebsd.org/mailman/listinfo/freebsd-current >To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org" > >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?414A02C2.4010009>