Created attachment 341 [details]
debug files
0 â gb1-li-lumberjack-004 ~ # rpm -qa|grep xfs
xfsprogs-3.2.0-0.10.alpha2.el7.x86_64
xfsdump-3.1.3-5.el7.x86_64
0 â gb1-li-lumberjack-004 ~ # uname -a
Linux gb1-li-lumberjack-004 3.10.0-123.el7.x86_64 #1 SMP Mon Jun 30 12:09:22
UTC 2014 x86_64 x86_64 x86_64 GNU/Linux
I have a box running CentOS 7.0.1406 on which the /opt partition has hung and
become unresponsive. Backing disk is a VMDK sitting on a HP 3Par 7200.
mount reports as rw:
0 â ~/xfs_issue # mount|grep opt
/dev/mapper/vg_root-opt on /opt type xfs
(rw,relatime,seclabel,attr2,inode64,noquota)
if you try and do any actions on the mount (even a lowly ls) the process hangs
in a blocking state. Nothing has been reported to dmesg, /var/log/messages or
the journalctl logs.
I got some of the trace-cmd logs, but not sure how helpful they'll be as the
mount is already in the hung state. I can't recreate the issue as I don't know
what causes it. It has happened a few times on some other of our machines, all
running higher IO applications (postgres, elasticsearch).
host was writing at ~12MB/s, reading at ~9MB/s at the time (based on our vmware
disk graphs).