An opportunity to investigate the fstab thing?

Feel free to talk about anything and everything in this board. Just don't post offensive topics that are meant to cause trouble with other members or are derogatory towards people of different genders, race, color, minors (this includes nudity and sex), politics or religion. Let's try to keep peace among the community and for visitors.

No spam on this or any other forums please! If you post advertisements on these forums, your account may be deleted.

Do not copy and paste entire or even up to half of someone else's words or articles into posts. Post only a few sentences or a paragraph and make sure to include a link back to original words or article. Otherwise it's copyright infringement.

You can talk about other distros here, but no MEPIS bashing. You can email the developer of MEPIS if you just want to say you dislike or hate MEPIS.
Message
Author
User avatar
Utopia
Forum Veteran
Forum Veteran
Posts: 3652
Joined: Sun Apr 29, 2007 11:53 am

An opportunity to investigate the fstab thing?

#1 Postby Utopia » Thu Nov 03, 2011 6:27 am

After reading DBeckett (Post # 287413)
except that wayward fstab thing has reappeared, where fstab has mistakes in it that weren't there before

Isn't this a great chance to find out if a package in the test repo is causing it?
At least one other user (Magnus Johansson) seems to have used the test repo when bitten by fstab. And the question was never asked (if the test repo was enabled) in the other cases.
I usually don't suggest how someone else shall spend his days, but if DBeckett has a day off this is a great opportunity.
Is fstab modified if the upgrade is repeated on a clean install? And is it possible to upgrade a few packages at a time to find the culprit, if there is one?
Henry

User avatar
DBeckett
Forum Veteran
Forum Veteran
Posts: 3070
Joined: Sat May 16, 2009 3:59 pm

Re: An opportunity to investigate the fstab thing?

#2 Postby DBeckett » Thu Nov 03, 2011 1:17 pm

.
I'm willing to test it, but it will take a while.

This fstab error is unusual. Partitions 13 and 14, below the "Dynamic entries" line, are listed as ext4, which they are not. This should throw an error if I try to mount them with Dolphin, but it does not—it works normally. Bizarre.
Gigabyte 990FXA-UD3, AMD FX-6100 hex-core, 3.3GHz, 8G, Radeon HD6570

User avatar
joany
Forum Veteran
Forum Veteran
Posts: 6123
Joined: Mon Feb 12, 2007 1:45 pm

Re: An opportunity to investigate the fstab thing?

#3 Postby joany » Thu Nov 03, 2011 1:20 pm

I don't think this has anything to do with installing packages in the test repository. I was bitten by the bug with a clean SM 11 installation and with nothing installed from that repository. Entries above the #Dynamic line in /etc/fstab would be removed intermittently, maybe every third time I booted.
MX-14; 3.12-0.bpo.1-686-pae kernel using 4GB RAM
2.4GHz AMD Athlon 4600+
NVidia GeForce 6150 LE; 304.121 Display Driver
You didn't slow down because you're old; you're old because you slowed down.

User avatar
Utopia
Forum Veteran
Forum Veteran
Posts: 3652
Joined: Sun Apr 29, 2007 11:53 am

Re: An opportunity to investigate the fstab thing?

#4 Postby Utopia » Thu Nov 03, 2011 1:37 pm

I don't think this has anything to do with installing packages in the test repository

If it doesn't, it would be nice to rule that out.
I was thinking of what Dropbox did and thought some app is trying to make the Mepis fstab look like a default debian fstab.
But this ext4 thing is a new twist, maybe more than one culprit-app must be present for this to happen?
I hope everyone who encounters fstab problems reports it on the forum.
Henry

User avatar
kmathern
Forum Veteran
Forum Veteran
Posts: 9169
Age: 58
Joined: Wed Jul 12, 2006 2:26 pm

Re: An opportunity to investigate the fstab thing?

#5 Postby kmathern » Thu Nov 03, 2011 1:42 pm

Utopia wrote:
I don't think this has anything to do with installing packages in the test repository

If it doesn't, it would be nice to rule that out.
I was thinking of what Dropbox did and thought some app is trying to make the Mepis fstab look like a default debian fstab.
But this ext4 thing is a new twist, maybe more than one culprit-app must be present for this to happen?
I hope everyone who encounters fstab problems reports it on the forum.
Henry

I also don't think installing packages from the test repo has anything to do with the fstab corruption.

Lately I've been installing quite a few packages from the test repo in the process of testing Stevo's KDE 4.6.5 upgrade, I haven't had any fstab corruption.

User avatar
Utopia
Forum Veteran
Forum Veteran
Posts: 3652
Joined: Sun Apr 29, 2007 11:53 am

Re: An opportunity to investigate the fstab thing?

#6 Postby Utopia » Thu Nov 03, 2011 1:52 pm

Maybe it's not the test repo. But the rarity of the error must depend on something.
It's not the usual stuff that everyone gets. Now we have a recent case and a situation when it happened. Can it be repeated? Worth investigating as the earlier fstab corruptions just happened without any clues.
If not any specific packages, could the updating process itself be involved?
Henry

User avatar
DBeckett
Forum Veteran
Forum Veteran
Posts: 3070
Joined: Sat May 16, 2009 3:59 pm

Re: An opportunity to investigate the fstab thing?

#7 Postby DBeckett » Thu Nov 03, 2011 1:58 pm

.
This is embarrassing. This fstab problem was my own fault. I had restored an old image from six months ago when partitions 13 and 14 were ext4. So please disregard this whole topic. I'm going to edit my post in the other thread.
Gigabyte 990FXA-UD3, AMD FX-6100 hex-core, 3.3GHz, 8G, Radeon HD6570

User avatar
timkb4cq
Forum Veteran
Forum Veteran
Posts: 3492
Joined: Wed Jul 12, 2006 4:05 pm

Re: An opportunity to investigate the fstab thing?

#8 Postby timkb4cq » Thu Nov 03, 2011 2:05 pm

I had it happen on an old M-8.5 install the other day. No test repo packages on that one.

I had to disable the buildfstab script on that machine because of repeated errors rewriting one line.
I know it's unusual practice, but for this particular machine one mount point must have a space in the pathname. Fstab has a way to handle that. The space is entered as the octal code for a space character \040 , so the line looks something like

Code: Select all

/dev/sda3 /home/user/My\040Data ext3 ,users,exec,noatime 0 0
Works like a charm until buildfstab rewrites it and removes the backslash.
It doesn't happen every time, but it happens a lot more frequently than the disappearing lines glitch. I suspect they're related.
MSI 970A-G43 MB, AMD FX-6300 (six core), 16GB RAM, GeForce 730, Samsung 850 EVO 250GB SSD, Seagate Barracuda XT 3TB

User avatar
Utopia
Forum Veteran
Forum Veteran
Posts: 3652
Joined: Sun Apr 29, 2007 11:53 am

Re: An opportunity to investigate the fstab thing?

#9 Postby Utopia » Thu Nov 03, 2011 2:13 pm

So please disregard this whole topic

That's sad. Thought we had the first clue, even if very small and vague.
I had it happen on an old M-8.5 install the other day.

A new clue. But so much more difficult to investigate.
The first one was more fitting for an amateur sleuth.
Henry

User avatar
DBeckett
Forum Veteran
Forum Veteran
Posts: 3070
Joined: Sat May 16, 2009 3:59 pm

Re: An opportunity to investigate the fstab thing?

#10 Postby DBeckett » Thu Nov 03, 2011 2:30 pm

Utopia wrote:That's sad. Thought we had the first clue, even if very small and vague.

Sorry to disappoint. :smile:
Gigabyte 990FXA-UD3, AMD FX-6100 hex-core, 3.3GHz, 8G, Radeon HD6570


Return to “General”

Who is online

Users browsing this forum: Yahoo [Bot] and 4 guests