Emergency UNIX and Linux Support

Hi All,

I have Solaris-10 configured with two non-global zones. All file-systems are mounted on global zone and data file-systems are mounted on non-global zone as lofs.
I have added 4 luns of 100 GB each and still not able to extend a file-system. This is production server, so I can not reboot it anyway.
I tried unmounting, exporting, importing and mounting it back, but still no luck.

---------- Post updated at 03:05 PM ---------- Previous update was at 12:20 PM ----------

I have rebooted the server and then able to fix the issue. Please consider this thread as closed

Could you please post the output of below commands?

Code:

zpool list
zpool status

You cannot increase the size of a ZFS file system by setting quota. That's really a very wrong idea. Quota is a way to set a hard limit on the maximum size of the ZFS file system.

How did you add the LUNs to the zpool? You need to check the total pool size with the "zpool list" command. If you want to see the underlying ZFS volume's size, use "zfs list" command. Data provided by df command for ZFS filesystems is kind of misleading due to the whole pooling concept.

We had enough new free disks added in pool. We are increasing file-system size by setting quota only. As you said, this is not right way to extend quota, what should be the correct way? Please suggest.

By default, zfs filesystem does auto-resizing of itself when the pool size is increased. That's the whole point of zfs, isn't it? But, in your case, as you want to restrict you filesystem to be of certain size after the pool size is increased, I would have followed the below steps to make sure things go just fine:

1. Remove the quota of the filesystem

Code:

zfs set quota=none pool/filesystem

2. Increase the size of the pool

Code:

zpool add blah blah

3. Now, restrict the filesystem size

Code:

zfs set quota=blah blah

It doesn't say anywhere in Solaris documentations, but I figured, it's always a good idea to unset the quota before setting it to a bigger number. Works every time!

Please post

Code:

zpool status
zpool list
zfs list -t all
df -k

output for us to understand the issue. Then please post also the quota settings you are using:

Code:

zfs get all | grep -i quota

The user asked for this thread to be closed and so it is.