Mailinglist Archive: opensuse-factory (464 mails)

< Previous Next >
Re: [opensuse-factory] 11.0 default filesystems again
  • From: Sid Boyce <sboyce@xxxxxxxxxxxxxxxx>
  • Date: Mon, 25 Feb 2008 21:54:58 +0000
  • Message-id: <47C33932.5050300@xxxxxxxxxxxxxxxx>
Carlos E. R. wrote:
Hash: SHA1

The Sunday 2008-02-24 at 08:47 -0000, Sid Boyce wrote:

Either that, or you are reporting the wrong units.

# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sdb1 401G 224G 178G 56% /
udev 2.0G 180K 2.0G 1% /dev
/dev/sdb2 56G 150M 56G 1% /boot
/dev/sda1 211G 31G 170G 16% /ftp

Uau. :-O

So, yes, you really waste 56 GB...

/dev/sdc2 is a swap partition not used currently.
I can alter this any time of course, but when it was done, I had rsync'd a failing IDE drive (ext3) over to a new SATA HD (/dev/sdb1) JFS and it refused to boot, so I just grabbed some space from swap to see if it would boot from a reiserfs partition.
I'm not likely to run short of space before I need to replace another failing HD.

Well, yes, some space, yes... but 100 MB is ample space. The rest you can put to use. Of course, you can symlink anything you like in there, no need to repartition.

> I have had solid lockups as recently as an hour ago (hardware I reckon) > and JFS comes up smiling, whereas an inadvertent power down of the other > box yesterday, 145G of ext3 caused a major headache.

But you miss the point: a 100 MB partition checks in seconds, regardles of
the filesystem you use. Even the old ext2.

And I'm not telling you to use a 145G part as ext3.

The point I was trying to make here is that I've had a problem with fsck.ext3 taking a very long time to repair a partition on a relative's box when it suffered an accidental power down.

That point has been understood messages ago :-)

It is a known issue.

The kernel people suffered it themselves, all their source store went down, and it took hours to fsck, I believe. The comment was that it have been faster to reformat and restore from backup.

Usually it is just a question of waiting calmy drinking your tea :-p

But, the other point is that a very small partition, of 100 MB, checks in seconds, allowing it to be of ext2 type, which is safer for a boot partition.

- -- Cheers,
Carlos E. R.

That's the time when the PHB paces around muttering about giving MS a call and tea drinking the ultimate act causing insult. Power does fail in most shops and a painful recovery doesn't help. I thought journalling was supposed to fix that. One other reason why I'd trust IBM, their mainframe filesystems always survive power outages, so I would expect JFS to have that kind of resilience designed in and data corruptions considered a disaster situation.
With regard to the statement that customers never asked for JFS, neither did they ask for reiserfs or ext3, they took what they were given.

Agreed on the 100MB, I kept thinking in GB for the boot partition. The biggie GB partitions still cause grief whichever way you deal with them.
Sid Boyce ... Hamradio License G3VBV, Licensed Private Pilot
Emeritus IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support Specialist, Cricket Coach
Microsoft Windows Free Zone - Linux used for all Computing Tasks

To unsubscribe, e-mail: opensuse-factory+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: opensuse-factory+help@xxxxxxxxxxxx

< Previous Next >