SOLVED: More bad luck-F19 booting into emergency mode

[whaites]

I'm having some more issues getting amahi running smoothly. I had finally set up Amahi, got squeeze server running and I think I had greyhole sorted. First I had some issue with one of the storage drives and none of the drives were writable, but fsck seems to have resolved that.

Now things are worse. Currently, fedora 19 only boots into emergency mode, so I think something is hosed with the system. I have fedora install on its own drive. Is there a utility I can run to sort this out, or at least figure out if it is a logical or physical problem with the drive? I believe I originally installed using Express CD 7.0, but then did some yum updates. Would I need that disk again (probably lost), or could I use the current version of Express CD to either recover or reinstall?

I *think* the data drives are currently ok, so I am hopeful there is a solution where I won't need to redo all the greyhole set-up again manually.

As always, thanks for the help.

[bigfoot65]

If you need to reinstall, recommend you the Amahi 7 Express DVD. Not sure what the problem could be, but may be drive related. Do all the drives mount on boot?

[whaites]

When it comes up in emergency mode, it really only gives me an option to look at the log - they were pretty huge, but I think I saw some red lines about the system or root mount. I think it has an issue mounting the main partition, or maybe it mounts but it won't let me run fsck on it.

Is the easiest path to recovery to just try a reinstall - assuming the disk isn't dead, or is the a repair util on the express disk I should try? Would I want to unplug all the data drives before doing express install?

[bigfoot65]

I would say reinstall would be quicker and probably best. There are no special utilities on the disc.

Recommend only the One drive be connected when doing the reinstall.

[ Post made via Android ]

[gorillazgr]

I don't know if this helps, (after such a long time) ,

but when I came across the same situation (the system started to boot only into emergency mode)
I found out that while the process of setting up greyhole ,
I had mistyped the extra lines in /etc/fstab regarding the greyhole pool drives....

so instead of the line:

UUID=a15857e8-ea22-4915-acac-a95618b39069 /var/hda/files/drives/drive3 ext4 defaults 1 2

I had the following:

UUID=a15857e8-ea22-4915-acac-a95618b39069 /var/hda/files/drives/drive3
ext4 defaults 1 2

PLEASE take note, that in the second example there are two lines instead of one!

[bigfoot65]

Yes that does make a difference. When it boots into emergency mode, it typically will tell you there is something wrong with the /etc/fstab. Guess in this case it did not. So all is good now I presume.

[whaites]

I ended up reinstalling. It was running fine for a while before booting to emergency mode. I think I had an issue with both the system drive and a data drive. The data drive has been replaced. I set the new fstab options as recommended in the greyhole site docs with an option to allow boot on drive error. I forget the syntax atm, hoping it will allow my server to run with a data drive issue in the future.

[ Post made via Android ]

[bigfoot65]

Cool. Would be good to know the syntax you added so we can update the wiki. I will see if I can find it.

[whaites]

Found it. From the Greyhole Mounts best practices page.

I used these options in place of defaults:

Code: Select all

noatime,data=writeback,nofail

Hoping the next fail on a data drive does not bomb my server.

[bigfoot65]

Cool. Could you please add this to the wiki? Good info that other users might find useful.