Skip to contentSkip to conversation

Kernel 2.2.6-ac1 problem

DISCUSSION root 5 participants 9 messages Apr 19, 1999 – Apr 22, 1999

lore
Share & tools
lore

root

author
#0 lore raw

I compiled the latest kernel from source and I created vmlinux with no problems. After installing it in my system folder on my mac partition and booting linux again, I get I/O errors somewhat randomly. The message that is displayed is "No space left on device". The file /var/log/messages conatins the following lines:

Apr 19 23:23:07 localhost kernel: EXT2-fs error (device ide0(3,4)):

ext2_new_block: Free blocks count corrupted for block group 13

I had the same problem when compiling and testing the 2.2.5 kernel as well, but I found no other references to a similar problem in the mailing lists and newsgroups. Can anyone help or offer advice?

-J. Robokos

System config:
Powerbase 603,
40MB Ram,
Maxtor IDE hard drive (Master),
Seagate IDE hard drive (Slave)

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]

#1 lore raw
Quoted 8 lines · root

I compiled the latest kernel from source and I created vmlinux with no problems. After installing it in my system folder on my mac partition and booting linux again, I get I/O errors somewhat randomly. The message that is displayed is "No space left on device". The file /var/log/messages conatins the following lines:

Apr 19 23:23:07 localhost kernel: EXT2-fs error (device ide0(3,4)):

ext2_new_block: Free blocks count corrupted for block group 13

Are you sure you use egcs-1.1.2 for compiling the kernel? egcs versions before 1.1 (or 1.1.1? I don't remember anymore) had this problem.

Franz.

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]

author
#2 lore raw
Quoted 8 lines · Franz Sirl
Quoted 6 lines · root

I compiled the latest kernel from source and I created vmlinux with no problems. After installing it in my system folder on my mac partition and booting linux again, I get I/O errors somewhat randomly. The message that is displayed is "No space left on device". The file /var/log/messages conatins the following lines:

Quoted 7 lines
Quoted 2 lines
Apr 19 23:23:07 localhost kernel: EXT2-fs error (device ide0(3,4)):

ext2_new_block: Free blocks count corrupted for block group 13

Are you sure you use egcs-1.1.2 for compiling the kernel? egcs versions before 1.1 (or 1.1.1? I don't remember anymore) had this problem.

Franz.

I re-compiled with Gary Thomas' egcs RPMS's and so far it seems the problem no longer exists. Thanks for the help.

-J. Robokos

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]

#3 lore raw
In reply to root
Quoted 8 lines · root

I compiled the latest kernel from source and I created vmlinux with no problems. After installing it in my system folder on my mac partition and booting linux again, I get I/O errors somewhat randomly. The message that is displayed is "No space left on device". The file /var/log/messages conatins the following lines:

Apr 19 23:23:07 localhost kernel: EXT2-fs error (device ide0(3,4)):

ext2_new_block: Free blocks count corrupted for block group 13

We've had similar problems with free block corruption under MkLinux (2.0.37). This may be a red herring, but I've only noticed the damage on the last partition mounted. You might try mounting a bogus partition after the others and see if it goes away. Just a wild thought. Anyway, the problem began when someone ported some changes from the current development sources into the 2.0.xx tree, between 2.0.33 and 2.0.36 (the former was fine, the latter has the problem. It might be easier to find in the context of 2.0.xx since many less things were changing....

Later,
David

David A. Gatwood Visit globegate's internet
[email protected] talker, Deep Space 36
http://globegate.utm.edu telnet globegate.utm.edu:9624

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]

#4 lore raw
Quoted 10 lines · David A. Gatwood
Quoted 8 lines · root

I compiled the latest kernel from source and I created vmlinux with no problems. After installing it in my system folder on my mac partition and booting linux again, I get I/O errors somewhat randomly. The message that is displayed is "No space left on device". The file /var/log/messages conatins the following lines:

Apr 19 23:23:07 localhost kernel: EXT2-fs error (device ide0(3,4)):

ext2_new_block: Free blocks count corrupted for block group 13

Quoted 8 lines

We've had similar problems with free block corruption under MkLinux (2.0.37). This may be a red herring, but I've only noticed the damage on the last partition mounted. You might try mounting a bogus partition after the others and see if it goes away. Just a wild thought. Anyway, the problem began when someone ported some changes from the current development sources into the 2.0.xx tree, between 2.0.33 and 2.0.36 (the former was fine, the latter has the problem. It might be easier to find in the context of 2.0.xx since many less things were changing....

Do you mean the change from big endian ext2fs to standard little endian ext2fs?

Greetings,

Geert

--
Geert Uytterhoeven [email protected]
Wavelets, Linux/{m68k~Amiga,PPC~CHRP} http://www.cs.kuleuven.ac.be/~geert/
Department of Computer Science -- Katholieke Universiteit Leuven -- Belgium

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]

#5 lore raw
Quoted 11 lines · Geert Uytterhoeven
Quoted 8 lines

We've had similar problems with free block corruption under MkLinux (2.0.37). This may be a red herring, but I've only noticed the damage on the last partition mounted. You might try mounting a bogus partition after the others and see if it goes away. Just a wild thought. Anyway, the problem began when someone ported some changes from the current development sources into the 2.0.xx tree, between 2.0.33 and 2.0.36 (the former was fine, the latter has the problem. It might be easier to find in the context of 2.0.xx since many less things were changing....

Do you mean the change from big endian ext2fs to standard little endian ext2fs?

No. I've been running little endian with linux servers running back to 2.0.28 or so, or shortly thereafter. 2.0.33 is stock DR3, MkLinux's current branch is 2.0.37 pre-patch whatever....

The related code I noticed changing was related to quota support. It doesn't appear to be in the ext2fs code changes, though, because I tried backdating all of those (other than simple formatting and notational changes0 to no avail. I'm thinking it might be in the general filesystem code.

Later,
David

David A. Gatwood Visit globegate's internet
[email protected] talker, Deep Space 36
http://globegate.utm.edu telnet globegate.utm.edu:9624

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]

#6 lore raw
Quoted 5 lines · David A. Gatwood

The related code I noticed changing was related to quota support. It doesn't appear to be in the ext2fs code changes, though, because I tried backdating all of those (other than simple formatting and notational changes0 to no avail. I'm thinking it might be in the general filesystem code.

any chance it could be the quota code? i have it enabled, but i don't have it setup (no quotas) and i get minor corruption when copying and deleting lots of files. does it happen with quotas disabled?

--
Tom Vier - 0x82B007A8
[email protected]
http://www.erols.com/thomassr/zero/

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]

author
#7 lore raw
In reply to David A. Gatwood
Quoted 20 lines · David A. Gatwood
Quoted 8 lines · root

I compiled the latest kernel from source and I created vmlinux with no problems. After installing it in my system folder on my mac partition and booting linux again, I get I/O errors somewhat randomly. The message that is displayed is "No space left on device". The file /var/log/messages conatins the following lines:

Apr 19 23:23:07 localhost kernel: EXT2-fs error (device ide0(3,4)):

ext2_new_block: Free blocks count corrupted for block group 13

We've had similar problems with free block corruption under MkLinux (2.0.37). This may be a red herring, but I've only noticed the damage on the last partition mounted. You might try mounting a bogus partition after the others and see if it goes away. Just a wild thought. Anyway, the problem began when someone ported some changes from the current development sources into the 2.0.xx tree, between 2.0.33 and 2.0.36 (the former was fine, the latter has the problem. It might be easier to find in the context of 2.0.xx since many less things were changing....

Quoted 6 lines

Later,
David

David A. Gatwood Visit globegate's internet
[email protected] talker, Deep Space 36
http://globegate.utm.edu telnet globegate.utm.edu:9624

I don't think the error is happening on my last partition. To verify that, I tried compiling a c program and I got an error (after a few tries) saying that ld failed writing to /tmp
The last few partitions of my hard drive are listed below:

...
7: Apple_UNIX_SVR2 opt 614400 @ 1823921 (300.0M)

8: Apple_UNIX_SVR2 home 102400 @ 2438321 ( 50.0M)

9: Apple_HFS MacOS 11527935 @ 2540721 ( 5.5G)

and the dirctory /tmp is on another partition on the same drive.

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]

#8 lore raw
Quoted 14 lines · J. Robokos
Quoted 8 lines

We've had similar problems with free block corruption under MkLinux (2.0.37). This may be a red herring, but I've only noticed the damage on the last partition mounted. You might try mounting a bogus partition after the others and see if it goes away. Just a wild thought. Anyway, the problem began when someone ported some changes from the current development sources into the 2.0.xx tree, between 2.0.33 and 2.0.36 (the former was fine, the latter has the problem. It might be easier to find in the context of 2.0.xx since many less things were changing....

I don't think the error is happening on my last partition. To verify that, I tried compiling a c program and I got an error (after a few tries) saying that ld failed writing to /tmp
The last few partitions of my hard drive are listed below:

Quoted 8 lines

...
7: Apple_UNIX_SVR2 opt 614400 @ 1823921 (300.0M)

8: Apple_UNIX_SVR2 home 102400 @ 2438321 ( 50.0M)

9: Apple_HFS MacOS 11527935 @ 2540721 ( 5.5G)

and the dirctory /tmp is on another partition on the same drive.

No, I mean the last one mounted, generally speaking, the last entry in fstab. It's also possible that these are two different problems.

David

David A. Gatwood Visit globegate's internet
[email protected] talker, Deep Space 36
http://globegate.utm.edu telnet globegate.utm.edu:9624

[[ This message was sent via the linuxppc-dev mailing list. Replies are ]] [[ not forced back to the list, so be sure to Cc linuxppc-dev if your ]] [[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting. ]]