zpool import taking a long time

31 views
Skip to first unread message

vajonam

unread,
Jul 30, 2019, 9:26:50 AM7/30/19
to EON ZFS Storage

I had a power failure, really a loose power lead fell out! so much for backup power! 

anyhow now one of my pools doesn't import correctly, it shows in the zfs list but the mount point is empty and not showing any of my data! 

the name of the pool is "gobi", 


  pool: gobi
 state: ONLINE
  scan: scrub canceled on Tue Jul 30 09:15:46 2019
config:

        NAME                       STATE     READ WRITE CKSUM
        gobi                       ONLINE       0     0     0
          mirror-0                 ONLINE       0     0     0
            c0t50014EE2B94B1268d0  ONLINE       0     0     0
            c0t50014EE2B94B2234d0  ONLINE       0     0     0
        logs
          c0t5000000000000000d0    ONLINE       0     0     0

errors: No known data errors


getafix:3:~#zfs list gobi
NAME   USED  AVAIL  REFER  MOUNTPOINT
gobi  2.88T   751G   152K  /gobi
getafix:4:~#zfs list gobi/media
NAME         USED  AVAIL  REFER  MOUNTPOINT
gobi/media  2.86T   751G  2.86T  /gobi/media

getafix:5:~#ls -l /gobi/media
total 0
getafix:6:~#


I am trying a zpool import -F gobi now, which is still running and lot of disk activity not sure when it will finish.. what are my options here? 

Thanks



'

Andre Lue

unread,
Jul 30, 2019, 1:33:24 PM7/30/19
to EON ZFS Storage on behalf of Donovan Kaardal
Not sure why its taking long. You can try truss on process or zfs dtrace script to try and see what its doing. 

I advise against a reboot or power it off. :)

Give it some time and lets see.




--
You received this message because you are subscribed to the Google Groups "EON ZFS Storage" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eonstorage+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eonstorage/f3577cd2-07dd-47fc-afb9-eaad50f737d1%40googlegroups.com.

vajonam

unread,
Aug 13, 2019, 9:38:00 AM8/13/19
to EON ZFS Storage
did a zpool import -f and it brought it back. I am still running the old eon nervous for the upgrade. I am going to get some upgraded hardware with more memory that should improve overall performance. 

To unsubscribe from this group and stop receiving emails from it, send an email to eonst...@googlegroups.com.

Andre Lue

unread,
Aug 13, 2019, 9:52:55 AM8/13/19
to EON ZFS Storage on behalf of Donovan Kaardal
Cool. Base ram requirement of new image is the same, ~300mb up to sky, so shouldnt be an issue.

To unsubscribe from this group and stop receiving emails from it, send an email to eonstorage+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eonstorage/4f27e2f2-bff0-4c50-83a3-f87ca3ed3694%40googlegroups.com.

vajonam

unread,
Dec 3, 2019, 6:55:05 AM12/3/19
to EON ZFS Storage
This happened again. I have a problem with my "/" partition filling up over some time, which results in me not being able to do much on the eon device, and needing to reboot. after such a reboot the "gobi" pool wont import, and, i will let it run fully for 24 hours.. i tried strace.. that didn't work what other command can I use to debug what its doing? you mentioned zfs dtrace script what is that? 



On Tuesday, August 13, 2019 at 9:52:55 AM UTC-4, dre2kse wrote:
Cool. Base ram requirement of new image is the same, ~300mb up to sky, so shouldnt be an issue.

vajonam

unread,
Dec 3, 2019, 1:20:18 PM12/3/19
to EON ZFS Storage
Looks like it coredumped and is rebooting after about 6 hours.. I dont have the core that was dumped. 

vajonam

unread,
Dec 3, 2019, 1:30:34 PM12/3/19
to EON ZFS Storage
running a truss zpool import gobi results ins ..


/2:     pread64(10, "\fB1BA\0\0\0\0\08813\0\0".., 262144, 262144) = 262144
/2:     pread64(10, "\0\0\0\0\0\0\0\0\0\0\0\0".., 262144, 0x000001FFFFF40000) = 262144
/2:     pread64(10, "\0\0\0\0\0\0\0\0\0\0\0\0".., 262144, 0x000001FFFFF80000) = 262144
/2:     close(10)                                       = 0
/2:     lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0xFFFFFFF7, 0x000000FF, 0x00000000) = 0xFFBFFEFF [0xFFFFFFFF]
/1:     lwp_park(0x00000000, 0)                         = 0
/5:     lwp_park(0x00000000, 0)                         = 0
/5:     lwp_unpark(4)                                   = 0
/5:     lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0xFFFFFFF7, 0x000000FF, 0x00000000) = 0xFFBFFEFF [0xFFFFFFFF]
/5:     lwp_exit()
/4:     lwp_park(0x00000000, 0)                         = 0
/4:     lwp_unpark(3)                                   = 0
/4:     lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0xFFFFFFF7, 0x000000FF, 0x00000000) = 0xFFBFFEFF [0xFFFFFFFF]
/4:     lwp_exit()
/3:     lwp_park(0x00000000, 0)                         = 0
/3:     lwp_unpark(2)                                   = 0
/3:     lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0xFFFFFFF7, 0x000000FF, 0x00000000) = 0xFFBFFEFF [0xFFFFFFFF]
/3:     lwp_exit()
/2:     lwp_park(0x00000000, 1)                         = 0
/2:     lwp_unpark(1)                                   = 0
/2:     lwp_sigmask(SIG_SETMASK, 0xFFBFFEFF, 0xFFFFFFF7, 0x000000FF, 0x00000000) = 0xFFBFFEFF [0xFFFFFFFF]
/2:     lwp_exit()
lwp_park(0x00000000, 5)                         = 0
close(7)                                        = 0
open("/dev/dsk/c0t50014EE2B94B2234d0s0", O_RDONLY) = 7
stat64("/lib/libdevid.so.1", 0x08042438)        = 0
resolvepath("/lib/libdevid.so.1", "/lib/libdevid.so.1", 1023) = 18
open("/lib/libdevid.so.1", O_RDONLY)            = 8
mmapobj(8, MMOBJ_INTERPRET, 0xFEC205C8, 0x080424A4, 0x00000000) = 0
close(8)                                        = 0
memcntl(0xFE9C0000, 4824, MC_ADVISE, MADV_WILLNEED, 0, 0) = 0
fstat(7, 0x08042E00)                            = 0
modctl(MODSIZEOF_DEVID, 0x01640080, 0x08042DFC, 0xFE9C14B1, 0xFED78F60) = 0
modctl(MODGETDEVID, 0x01640080, 0x00000014, 0x080CE1A8, 0xFED78F60) = 0
fstat(7, 0x08042E00)                            = 0
modctl(MODSIZEOF_MINORNAME, 0x01640080, 0x00006000, 0x08042DFC, 0xFED78F68) = 0
modctl(MODGETMINORNAME, 0x01640080, 0x00006000, 0x00000002, 0x0808FF68) = 0
close(7)                                        = 0
ioctl(3, ZFS_IOC_POOL_STATS, 0x08041DE0)        Err#2 ENOENT
ioctl(3, ZFS_IOC_POOL_TRYIMPORT, 0x08041E50)    Err#12 ENOMEM
ioctl(3, ZFS_IOC_POOL_TRYIMPORT, 0x08041E50)    = 0
sysinfo(SI_HW_SERIAL, "311901069", 11)          = 10
open("/usr/lib/locale/en_CA.UTF-8/LC_MESSAGES/SUNW_OST_OSLIB.mo", O_RDONLY) Err#2 ENOENT

vajonam

unread,
Dec 3, 2019, 1:35:55 PM12/3/19
to EON ZFS Storage

Here is the log file... 
screenlog.0

Andre Lue

unread,
Dec 3, 2019, 3:06:02 PM12/3/19
to EON ZFS Storage on behalf of Donovan Kaardal
/ filling up means you have something logging to image/memory vs disk/pool. Need to fix this. I also need to upload updated image for you to test based on omni os. 

Core prob failed because space issue related to above explanation.

On Tue, Dec 3, 2019, 6:55 AM vajonam via EON ZFS Storage <eonstorage+APn2wQeaGveeVtAFGo6St...@googlegroups.com> wrote:
This happened again. I have a problem with my "/" partition filling up over some time, which results in me not being able to do much on the eon device, and needing to reboot. after such a reboot the "gobi" pool wont import, and, i will let it run fully for 24 hours.. i tried strace.. that didn't work what other command can I use to debug what its doing? you mentioned zfs dtrace script what is that? 



On Tuesday, August 13, 2019 at 9:52:55 AM UTC-4, dre2kse wrote:
Cool. Base ram requirement of new image is the same, ~300mb up to sky, so shouldnt be an issue.

To unsubscribe from this group and stop receiving emails from it, send an email to eonstorage+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eonstorage/cc660294-a912-423f-99f3-c3aff5a2863d%40googlegroups.com.

Manojav Sridhar

unread,
Dec 3, 2019, 8:07:04 PM12/3/19
to EON ZFS Storage on behalf of dre2kse
Understood. I am trying to import one disk from this mirrored pool  into a Linux box for now so I get access to the files atleast. That import is now been running for sometime (4 hours) 



On Dec 3, 2019, at 3:06 PM, EON ZFS Storage on behalf of dre2kse <eonst...@googlegroups.com> wrote:



vajonam

unread,
Dec 4, 2019, 7:12:37 AM12/4/19
to EON ZFS Storage
I was able to do an import on a Linux box and then move the disk back to eon. That then imported quickly and the 2nd disk was resilvered.

The / file system filling up is still an issue. I will keep and eye on it and see what’s causing it to grow.

Manojav Sridhar

unread,
Dec 4, 2019, 8:16:33 AM12/4/19
to EON ZFS Storage on behalf of vajonam
I am ready to try the new omnios based images, can you include the mpt_sas and mpt2_sas drivers in the base image :-) thanks. 

On Wed, Dec 4, 2019 at 7:12 AM vajonam via EON ZFS Storage <eonstorage+APn2wQeaGveeVtAFGo6St...@googlegroups.com> wrote:
I was able to do an import on a Linux box and then move the disk back to eon. That then imported quickly and the 2nd disk was resilvered.

The / file system filling up is still an issue. I will keep and eye on it and see what’s causing it to grow.

--
You received this message because you are subscribed to the Google Groups "EON ZFS Storage" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eonstorage+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eonstorage/05c86361-387e-4f90-a9ac-986968281b69%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages