I attached a disk to an existing pool which contained just one drive so that it became mirrored and while that was resilvering the system hung, no remote login exiting logins did not respond. The system was aborted and sync run but during the boot the system wedged again. Booting milestone none and letting it come up from there I found all was well until zfs mount -a was run and then this would hang (10 minutes before I gave up). I booted milestone none again and exported the zpool and then the system would go into multi user. However unfortunately the dump was not complete. I''ve just tried to import the pool again and the system has hung again. I''ll leave it an hour this time. Is this a known behaviour? The pool should be o.k. as it still has one side of the mirror which should still be good. -- Chris Gerhard. __o __o __o PTS in Europe _`\<,`\<,`\<,_ Sun Microsystems Limited (*)/---/---/ (*) Phone: +44 (0) 1252 426033 (ext 26033) ----------------------------------------------------------- http://blogs.sun.com/chrisg ----------------------------------------------------------- NOTICE: This email message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3186 bytes Desc: S/MIME Cryptographic Signature URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20051123/9e0660d1/attachment.bin>
Chris Gerhard wrote:> > I attached a disk to an existing pool which contained just one drive so > that it became mirrored and while that was resilvering the system hung, > no remote login exiting logins did not respond. The system was aborted > and sync run but during the boot the system wedged again. > > Booting milestone none and letting it come up from there I found all was > well until zfs mount -a was run and then this would hang (10 minutes > before I gave up). > > I booted milestone none again and exported the zpool and then the system > would go into multi user. However unfortunately the dump was not complete. > > I''ve just tried to import the pool again and the system has hung again. > I''ll leave it an hour this time. > > Is this a known behaviour? The pool should be o.k. as it still has one > side of the mirror which should still be good.Running vmstat 1 on the console while it tried to do this stops at this point: 0 0 0 0 66296 30 153 24 0 877 0 450050 117 3 0 251 590 1178 6420 0 43 57 0 0 0 0 51584 6 123 40 0 229 0 723117 110 5 0 158 516 1208 3921 0 42 58 kthr memory page disk faults cpu r b w swap free re mf pi po fr de sr s0 s1 s4 s6 in sy cs us sy id 0 0 0 0 37184 38 270 326 0 56 0 858835 87 18 0 192 551 1193 5270 0 38 62 0 0 0 0 30464 32 145 8 0 413 0 1015689 120 1 0 156 502 1192 3893 0 51 49 At this point all the windows freeze. The previous crash dumps failed due to the dump device not being big enough. sigh. Time for kmdb -- Chris Gerhard. __o __o __o PTS in Europe _`\<,`\<,`\<,_ Sun Microsystems Limited (*)/---/---/ (*) Phone: +44 (0) 1252 426033 (ext 26033) ----------------------------------------------------------- http://blogs.sun.com/chrisg ----------------------------------------------------------- NOTICE: This email message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.