Frustrating lack of answers/ideas to Internal server error

[bigfoot65]

That shouldn't cause any problems unless you moved your web-apps to the new drive. Can you give us some details on your issue?

[mal]

That shouldn't cause any problems unless you moved your web-apps to the new drive. Can you give us some details on your issue?

It's a bit of a long tale....

I had an HDA in which I had two software RAIDed 500Gb drives. These started to fill so I made the decision to look at Greyhole as an expansion strategy. I also bought a 2TB drive. The plan being to have the 2TB & 2 * 500GB as a storage cluster.

I built the system on the 2TB drive but had lots of stability problems. Machine locked up all the time requiring a reboot daily. Eventually it seemed to settle so I added the first 500GB drive as part of Greyhole and asked to share pics and music on it.

The stability then went out of the window with random kernal crashes. I noticed the problem described above at boot up with the starting udev init error 255.

I decided to rebuild the server only this time I built it on the 500GB disk and decided I would install the 2TB disk as a second disk and store all the /var/hda/files on there. When only the 500GB drive was in the system there was no issue with the starting udev but since adding the second disk I get it just the same.

I am currently restoring data onto the server and it's been stable for 12 hours but I have still to install my apps as yet.

Any thoughts you have would be most welcome (if more info needed then please ask)
Thanks
Mal

[bigfoot65]

Might help if you post your /etc/fstab file. Chances are there may be a setting that is not quite right. We can then see what we can do to help you.

[mal]

Might help if you post your /etc/fstab file. Chances are there may be a setting that is not quite right. We can then see what we can do to help you.

Here you go .....

Code: Select all

# # /etc/fstab # Created by anaconda on Tue Nov 30 20:23:16 2010 # # Accessible filesystems, by reference, are maintained under '/dev/disk' # See man pages fstab(5), findfs(8), mount(8) and/or blkid(8) for more info # /dev/mapper/vg_tackroom-lv_root / ext4 defaults 1 1 UUID=48ab3f83-44e2-41e5-992b-f27633564c9b /boot ext4 defaults 1 2 /dev/mapper/vg_tackroom-lv_swap swap swap defaults 0 0 tmpfs /dev/shm tmpfs defaults 0 0 devpts /dev/pts devpts gid=5,mode=620 0 0 sysfs /sys sysfs defaults 0 0 proc /proc proc defaults 0 0 UUID=09904373-52ee-4df9-a334-a808b1e4fec1 /var/hda/files ext4 defaults 1 2

I added the last line and hope you can figure something from that
Thanks
Mal

[mal]

Apologies for nudging my own post .....

but I am starting to think that somehow the LVM could be causing me the grief here? Does that sound reasonable? Should I attempt a reinstall but turn off the LVM during the install?

What would be the best way of building the system? I have 1*2TB 3*500GB and a number of smaller disks I can use.

My plan is to have a 500GB Primary for the system and "home" directories with a 2TB and the other 2*500GB available to put in a Greyhole pool. I want at least 2 copies of music, at least 3 copies of photos with movie rips only needed the one time. I'd like the home directories backed up too ...can I use Crashplan to do that .... or should I minimise the actual amount stored in the Home Directory

So many questions but I can't seem to figure out the best approach?

Would I be wise to install on the 500GB first then add the 2TB then the other 500GB, one at a time?

Your guidance would be most welcome.
Mal

[bigfoot65]

You are probably right. What I did is unplugged my extra drives, installed Fedora and Amahi. Then I shutdown, connected my extra drive and ran hda-diskmount.

Everything worked fine then. Since the wiki and that script are for adding a drive, it does not take into account drives that are already mounted in Fedora and sucked into LVM.

That is the route I would try. You can turn LVM off or just leave it alone since the extra drives will not be recognized anyway during install.

Hope that helps.

[mal]

You are probably right. What I did is unplugged my extra drives, installed Fedora and Amahi. Then I shutdown, connected my extra drive and ran hda-diskmount.

Everything worked fine then. Since the wiki and that script are for adding a drive, it does not take into account drives that are already mounted in Fedora and sucked into LVM.

That is the route I would try. You can turn LVM off or just leave it alone since the extra drives will not be recognized anyway during install.

Hope that helps.

I really appreciate you taking the time to reply but since I don't know you well, I cannot figure out what you are advising me to do - the way you have written it is very ambiguous!! allow the lvm to default or not ? Follow the wiki or not?

I think I will try without LVM as a first step and see where that takes me!

I'll report back after the weekend.
Thanks
Mal

[bigfoot65]

Sorry if I was not clear. What I mean is to disconnect all your drives except the one for the OS. Install Fedora 12 and Amahi. Then once it's installed and working, shutdown the box and reconnect your drives.

Now you can follow the wiki for the hda-diskmount script. Hope that clears things up.

[mal]

Sorry if I was not clear. What I mean is to disconnect all your drives except the one for the OS. Install Fedora 12 and Amahi. Then once it's installed and working, shutdown the box and reconnect your drives.

Now you can follow the wiki for the hda-diskmount script. Hope that clears things up.

I rebuilt the server over the weekend .... I had made a mistake of not properly partitioning the disk on my previous builds .... so I now have /, /var and /home in different partitions on the main disk , added the second disk but only added it to the pool rather than moved /var/hda/files onto it.

I still have the error on boot up but I think that's more to do with the OS interaction with the disk SMART data.

Stability reigns supreme again .... welcome back HDA
Mal