On 05-Sep-2017 12:00, Joe Landman wrote:
On 09/05/2017 01:28 PM, mathog wrote:
Short form:
Is there a method to remount the logical volumes writable
that does not require a reboot?
Generally the FW would write lock it. A
mount -o remount,rw $path
may not clear this. I've found that I need to often do something akin
to
echo "- - -" > /sys/class/scsi_host/host0/scan
for each scsi host bus. Another thing to try is to remove the driver
and modprobe it again. However, as your /boot and / are on it, this
probably won't work well.
/boot wasn't, only / was.
Is there a difference between "mount -o remount" and "mount -a" if the
partions/logical volumes are already mounted "ro"? The system could
read
/etc/fstab which indicated that the mount should be rw. It isn't clear
to me from the man page what mount is supposed to do in that case.
Also, can anybody suggest what could possibly differ between
RAID degraded=1, logical volumes mounted ro, 1 disk failed, 1 disk
offline
and
RAID degraded=1, logical volumes mounted ro, RAID 1 disk failed, all
others online
such that "cat" and "more" did not work in the former but did work in
the latter, while "ls" and "megacli" worked in both?
At this point I have that just "happy to be walking away" feeling about
the whole incident.
Thanks,
David Mathog
mat...@caltech.edu
Manager, Sequence Analysis Facility, Biology Division, Caltech
_______________________________________________
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
