Hello, After updating to 11.1 my home server can't boot with errors like (ada0:ata2:0:0:0): WRITE_DMA48. ACB: 35 00 50 29 10 40 6c 00 0c 00 (ada0:ata2:0:0:0): CAM status: Command timeout (ada0:ata2:0:0:0): Retrying command for all 6 sata hdd(stripe from 2 raidz) ACB different from boot to boot Sometimes it even can boot, but in few minutes will hang with same errors. Hardware: Supermicro X8DTN+-F / 6xWD1502FYPS-02W3B0 /2xE5649 HDDs connected to sata ports on baseboard. If I add hint.ata.2.mode=PIO4 hint.ata.3.mode=PIO4 hint.ata.4.mode=PIO4 hint.ata.5.mode=PIO4 to device.hints I'm able to boot but performance of IO becomes really disappointing. Also, if I add something like "find / -name something" to zfs rc script, than boots fine. If I roll back system to 11.0 all works fine again. Any advise on debuging of this issue? Also raised bug report some time ago for this issue https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=221704
On Wed, Sep 20, 2017 at 11:06:11PM +0000, freebsd-stable List wrote:> Sometimes it even can boot, but in few minutes will hang with same errors. > > Hardware: Supermicro X8DTN+-F / 6xWD1502FYPS-02W3B0 /2xE5649 > HDDs connected to sata ports on baseboard. > If I add > hint.ata.2.mode=PIO4 > hint.ata.3.mode=PIO4 > hint.ata.4.mode=PIO4 > hint.ata.5.mode=PIO4 > to device.hints I'm able to boot but performance of IO becomes really disappointing. > Also, if I add something like "find / -name something" to zfs rc script, than boots fine.Boots fine but hangs in a few minutes? Or all is good to go? Here is a thught: it could be the find / ... command is causing enough delay to let some hardware "settle down" and behave more reasonably. Can you replace the find cmd with an equivalent sleep(1) command of like duration and get the same results?> If I roll back system to 11.0 all works fine again.Seemingly minor deltas in the system could have had the needed delay as an implicit part of the system... Just my 2c -- Keith Hellman #include <disclaimer.h> khellman at mcprogramming.com from disclaimer import standard khellman at mines.edu gpg key 9FCF40FD freenode.net as mrtuple "The First Python function ever written (takes place in the Garden of Eden)" Guido sayeth "I will write def foo():" "Hmm, I could use an import, or two", Satan said, in a whirl, "Why not write it in Perl?", and the second function ever written - def foo_you(): -- Python Limmerick Contest submission by cappy2112 http://groups-beta.google.com/group/comp.lang.python/browse_thread/thread/d7a780beaff2e88a/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 163 bytes Desc: not available URL: <http://lists.freebsd.org/pipermail/freebsd-stable/attachments/20170921/9088c39d/attachment.sig>
22 ???????? 2017 ?., 02:35, "Keith Hellman" <khellman at mcprogramming.com> ???????:> On Wed, Sep 20, 2017 at 11:06:11PM +0000, freebsd-stable List wrote: > >> Sometimes it even can boot, but in few minutes will hang with same errors. >> >> Hardware: Supermicro X8DTN+-F / 6xWD1502FYPS-02W3B0 /2xE5649 >> HDDs connected to sata ports on baseboard. >> If I add >> hint.ata.2.mode=PIO4 >> hint.ata.3.mode=PIO4 >> hint.ata.4.mode=PIO4 >> hint.ata.5.mode=PIO4 >> to device.hints I'm able to boot but performance of IO becomes really disappointing. >> Also, if I add something like "find / -name something" to zfs rc script, than boots fine. > > Boots fine but hangs in a few minutes? Or all is good to go? Here is a > thught: it could be the find / ... command is causing enough delay to > let some hardware "settle down" and behave more reasonably. Can you > replace the find cmd with an equivalent sleep(1) command of like > duration and get the same results? > >> If I roll back system to 11.0 all works fine again. > > Seemingly minor deltas in the system could have had the needed delay > as an implicit part of the system... > > Just my 2cNope, it can't boot fine at all, or init will get segfault or ttys or just system becomes unresponsive and I can't even login, so this heppens during start of services. I've also tried to disable all "non-base" services, but without success. As for trying just sleep - it doesn't helps. Seems that find helps because of caching by system triggered by find.> -- > Keith Hellman #include <disclaimer.h> > khellman at mcprogramming.com from disclaimer import standard > khellman at mines.edu gpg key 9FCF40FD > freenode.net as mrtuple > > "The First Python function ever written (takes place in the Garden of Eden)" > > Guido sayeth "I will write def foo():" > "Hmm, I could use an import, or two", > Satan said, in a whirl, "Why not write it in Perl?", > and the second function ever written - def foo_you(): > > -- Python Limmerick Contest submission by cappy2112 > http://groups-beta.google.com/group/comp.lang.python/browse_thread/thread/d7a780beaff2e88a