Aloha list,

I recently installed Openfiler release 2.3 from livecd into a VMware ESX VM and have been very pleased with the web interface and features. I created a 8GB .vmdk file to store the /boot, /, and swap partitions. I mapped a 1TB RDM (raw LUN) to the virtual machine where I intend on keeping all of my user data. Rather than create one large filesystem, I created a 1TB partition to be a physical volume and have made a volume group which contained a 300GB logical volume. Things were working great until I ran out of space in my logical volume. I saw that the web-ui allows me to extend the logical volume, but not the ext3 filesystem. I went ahead and extended the lv to include another 100GB of extents.

I've done some searching through this list's archive and it seems that resizing filesystems used to be supported via the webui
http://article.gmane.org/gmane.linux.distributions.openfiler.user/502/match=ext3+resize

Is this still the case and I'm not looking in the right places? or is filesystem resizing only intended to be done from the command-line? Ultimately I hacked through ext2online, ext2resize, and resize2fs in order to take advantage of the extents that were added to my logical volume, but not recognized by the filesystem. On a second note, I wasn't able to use ext2online because of errors about "inode table has offset <something> when it should be <somethingelse>". I ran fsck.ext3 with the filesystem unmounted, which didn't help. Ultimately my command-line wanderings led me to a combination of "e2fsck -f /dev/vg/lv" and "resize2fs /dev/vg/lv" while the volume was offline.

I'd like to know if resizing ext3 filesystems is supposed to be supported while online. Maybe I just have a screwy SAN or something which prevented me from doing so. I'd also like to know if resizing filesystems is possible via the web-interface.

Thanks in advance,
John
_______________________________________________
Openfiler-users mailing list
[email protected]
https://lists.openfiler.com/mailman/listinfo/openfiler-users

Reply via email to