Hi,
(Solaris 9 9/05 HW / VxVM 5.0-2006-05-11a).
Last night me made massive mirror creations/synchronisations ("vxassist
mirror......") from an EMC Clarriion to a HP XP 24K.
There were approximately 80 vxassist/vxplex commands running in parrallel.
During the synchronisation, all VxVM commands (vxprint, vxtask, etc.)
were very slowed down, making the VX process monitoring completely
unusable : for instance some vxprint took 20 minutes to complete,
vxtask 25 minutes, etc. Some volume creation issued during the sync
process did not complete after 3 hours (I aborted them).
Also 3 of 80 vxplex failed with the following messages :
VxVM vxplex ERROR V-5-1-10128 Transaction locks timed out
VxVM vxassist ERROR V-5-1-502 Could not attach new mirror(s) to volume XXXXX
I tried to truss the vxconfigd daemon, it spent most of its time waiting
in some blocking ioctl command "ioctl(1,xxxx,xxxx)", so it seems the
blocking factor was kernel-side with probably some locking mechanism.
Note that except for the 3 vxplex that failed, the overall mirroring
process was completed and relatively fast.
Global system response was also satisfying.
Is this an usual behaviour to have vxvconfigd almost hanging during
heavy plex resync load or is it some sort of bug ?
Thanks in advance for your opinion or assistance.
--
Sebastien DAUBIGNE
sebastien.daubi...@atosorigin.com - +33(0)5.57.89.31.09
AtosOrigin Infogerance - AIS/D1/SudOuest/Bordeaux/IS-Unix
_______________________________________________
Veritas-vx maillist - Veritas-vx@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-vx
