1 |
Ramon van Alteren wrote: |
2 |
|
3 |
> Michiel de Bruijne wrote: |
4 |
> |
5 |
>> Where you able to resolve this? |
6 |
>> |
7 |
>> |
8 |
> Yep I have been able to resolve this by switching to the userspace |
9 |
> nfsd implementation |
10 |
> It appears to be a kernel bug with regard to the combination of LVM2 + |
11 |
> NFS-kernelspace + 9Tb |
12 |
|
13 |
It turned out to be another bug related to reiserfs. |
14 |
|
15 |
For those interested more data is here [1]: |
16 |
http://sourceforge.net/mailarchive/forum.php? |
17 |
forum_id=4930&max_rows=25&style=flat&viewmonth=200511&viewday=1 |
18 |
|
19 |
and here [2]: |
20 |
http://marc.theaimsgroup.com/?l=reiserfs&m=112255940804474&w=2 |
21 |
|
22 |
In short: Although reiserfs reports in it's FAQ that it supports |
23 |
filesystems up to 16Tb with the default options (4K blocks) it supports |
24 |
only 8Tb. It doesn't fail however and appears to work correctly when |
25 |
using the local filesystem, the problems start showing up when using NFS. |
26 |
|
27 |
From the post on [2] this is caused by a bug in reiserfsprogs that |
28 |
doesn't fully check the blockdescriptor when formatting. Apart from that |
29 |
the FAQ entry stating that reiserfs supports filesystems up to 16Tb is |
30 |
just plain wrong. |
31 |
|
32 |
I fixed the problem by using another filesystem. Based on comments on |
33 |
the nfs-ml and the excellent article in last months linuxgazette on |
34 |
filesystem performance we switched to jfs. |
35 |
So far it's holding up very good. We haven't seen the problems |
36 |
reappearing with files in excess of 1Gb |
37 |
|
38 |
Will be stress-testing nfs probably this week. If anybody is interested |
39 |
I could post the data ? |
40 |
|
41 |
Regards, |
42 |
|
43 |
Ramon |
44 |
|
45 |
-- |
46 |
To be stupid and selfish and to have good health are the three requirements for happiness, though if stupidity is lacking, the others are useless. |
47 |
|
48 |
Gustave Flaubert |
49 |
|
50 |
|
51 |
|
52 |
|
53 |
-- |
54 |
gentoo-server@g.o mailing list |