I need to sync ~3TB data from NFS share back to my crypto ZFS pool. I'm unable to do so as system gets frozen when ~400GB is copied to the dataset. If I do a hard reset (nothing else is possible) and resume copying, it again gets frozen after ~400GB copied.
My HW configuration:
Intel S1200BT with Xeon(R) CPU E3-1240 V2 @ 3.40
32GB ECC ram
LSI 9211-8i IT fw with 6 WD RED 2TB disks + Intel 120GB SSD
FreeBSD amd64 9.2-RELEASE r256438. Kernel is a custom built one (GENERIC + PF + crypto).
ZFS setup:
6x2TB disks are in radiz2 pool (pool portal) with SSD acting as a read cache. I've created 4TB zvol that is used with geli to create fully encrypted pool (pool temple). This configuration was working fine, but I did hit the issue described here: http://forums.freebsd.org/viewtopic.php?f=48&t=45814 and I had to recreate it all from the scratch.
Situation on system is as follows:
I'm using rsync to sync the data to the temple dataset:
System gets to the state where I can still ping it and I even get the version string from sshd when I try to telnet to the machine on port 22. I can't login on console. Any active window with opened terminal to this server works only to show the prompt (i.e. when I hit the enter with no command). If I enter a command, it gets hung.
After I reboot the server and mount back the dataset, I see data stopped after ~400GB. If I resume the rsync, system gets frozen again after 400GB (800GB synced now in total).
There's absolutely no log in syslog, disks show no errors at all. Kernel is configured with the watchdog, but due to the state it gets it never gets triggered.
I was wondering if anybody can help give me a hint how to, somehow, initiate a panic in this state so I have a dump available.
My HW configuration:
Intel S1200BT with Xeon(R) CPU E3-1240 V2 @ 3.40
32GB ECC ram
LSI 9211-8i IT fw with 6 WD RED 2TB disks + Intel 120GB SSD
FreeBSD amd64 9.2-RELEASE r256438. Kernel is a custom built one (GENERIC + PF + crypto).
ZFS setup:
6x2TB disks are in radiz2 pool (pool portal) with SSD acting as a read cache. I've created 4TB zvol that is used with geli to create fully encrypted pool (pool temple). This configuration was working fine, but I did hit the issue described here: http://forums.freebsd.org/viewtopic.php?f=48&t=45814 and I had to recreate it all from the scratch.
Situation on system is as follows:
Code:
# zpool status portal
pool: portal
state: ONLINE
scan: scrub repaired 0 in 28h50m with 0 errors on Sun Apr 6 08:07:45 2014
config:
NAME STATE READ WRITE CKSUM
portal ONLINE 0 0 0
raidz2-0 ONLINE 0 0 0
da0 ONLINE 0 0 0
da1 ONLINE 0 0 0
da2 ONLINE 0 0 0
da4 ONLINE 0 0 0
da5 ONLINE 0 0 0
da7 ONLINE 0 0 0
cache
da6 ONLINE 0 0 0
errors: No known data errors
#
# zfs list portal/zvols/bolt00
NAME USED AVAIL REFER MOUNTPOINT
portal/zvols/bolt00 4.01T 6.26T 837G -
#
# geli status
Name Status Components
zvol/portal/zvols/bolt00.eli ACTIVE zvol/portal/zvols/bolt00
#
# zpool status temple
pool: temple
state: ONLINE
scan: none requested
config:
NAME STATE READ WRITE CKSUM
temple ONLINE 0 0 0
zvol/portal/zvols/bolt00.eli ONLINE 0 0 0
errors: No known data errors
#
Code:
rsync -avH /path/to/nfs/share /path/to/my/temple/dataset
System gets to the state where I can still ping it and I even get the version string from sshd when I try to telnet to the machine on port 22. I can't login on console. Any active window with opened terminal to this server works only to show the prompt (i.e. when I hit the enter with no command). If I enter a command, it gets hung.
After I reboot the server and mount back the dataset, I see data stopped after ~400GB. If I resume the rsync, system gets frozen again after 400GB (800GB synced now in total).
There's absolutely no log in syslog, disks show no errors at all. Kernel is configured with the watchdog, but due to the state it gets it never gets triggered.
I was wondering if anybody can help give me a hint how to, somehow, initiate a panic in this state so I have a dump available.