[opensuse-kernel] Btrfs oops in 11.4
All, I just had a btrfs oops with a default 11,4 kernel. I haven't even tried a update kernel yet since I wasn't sure anyone would care. Unless advised otherwise, my plan is to try the same test with a Tumbleweed kernel and open a bugzilla if and only if I get the oops there too. Greg -- Sent from my mobile device Greg Freemyer Head of EDD Tape Extraction and Processing team Litigation Triage Solutions Specialist http://www.linkedin.com/in/gregfreemyer CNN/TruTV Aired Forensic Imaging Demo - http://insession.blogs.cnn.com/2010/03/23/how-computer-evidence-gets-retriev... The Norcross Group The Intersection of Evidence & Technology http://www.norcrossgroup.com -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 07/08/2011 02:18 PM, Greg Freemyer wrote:
All,
I just had a btrfs oops with a default 11,4 kernel.
I haven't even tried a update kernel yet since I wasn't sure anyone would care.
Unless advised otherwise, my plan is to try the same test with a Tumbleweed kernel and open a bugzilla if and only if I get the oops there too.
And the award for the most useless bug report goes to... ;) Seriously, though, btrfs oops are tough to track because error handling is currently implemented by BUG_ON, so the details are important. - -Jeff - -- Jeff Mahoney SUSE Labs -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org/ iEYEARECAAYFAk4XaLMACgkQLPWxlyuTD7IjCwCeMw8VIKEMJv/zbHLOR849BJDh 8YMAoJGTQe3xn92EzLRQ6Q54wiIAuKbO =T02x -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org
On 07/08/2011 10:29 PM, Jeff Mahoney wrote:
On 07/08/2011 02:18 PM, Greg Freemyer wrote:
All,
I just had a btrfs oops with a default 11,4 kernel.
I haven't even tried a update kernel yet since I wasn't sure anyone would care.
Unless advised otherwise, my plan is to try the same test with a Tumbleweed kernel and open a bugzilla if and only if I get the oops there too.
And the award for the most useless bug report goes to... ;)
Seriously, though, btrfs oops are tough to track because error handling is currently implemented by BUG_ON, so the details are important.
-Jeff
Greg can you detail what and how we can try to reproduce it. I've an (non critical) btrfs lvm that can be crash, or try to with the kernel 3 under factory. -- Bruno Friedmann Ioda-Net Sàrl www.ioda-net.ch openSUSE Member & Ambassador GPG KEY : D5C9B751C4653227 irc: tigerfoot
On Fri, Jul 8, 2011 at 5:10 PM, Bruno Friedmann <bruno@ioda-net.ch> wrote:
On 07/08/2011 10:29 PM, Jeff Mahoney wrote:
On 07/08/2011 02:18 PM, Greg Freemyer wrote:
All,
I just had a btrfs oops with a default 11,4 kernel.
I haven't even tried a update kernel yet since I wasn't sure anyone would care.
Unless advised otherwise, my plan is to try the same test with a Tumbleweed kernel and open a bugzilla if and only if I get the oops there too.
And the award for the most useless bug report goes to... ;)
Seriously, though, btrfs oops are tough to track because error handling is currently implemented by BUG_ON, so the details are important.
-Jeff
Greg can you detail what and how we can try to reproduce it. I've an (non critical) btrfs lvm that can be crash, or try to with the kernel 3 under factory.
--
Bruno Friedmann
Bruno, I'll try kernel 3.0 As for procedure, I've got a new xfstest that we're trying to develop. Works somewhat on one platform. Oopses on another. (I don't yet know what the difference is yet. Possibly 32bit vs. 64bit.) If you have the xfstests infrastructure setup, I could just send you our new test. It has known errors, but causing a oops I consider a success! If you don't have xfstests, I hope to create a wiki page that describes the specific install / config steps for openSUSE. I might get to that this weekend. Greg -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org
On Fri, Jul 8, 2011 at 4:29 PM, Jeff Mahoney <jeffm@suse.de> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 07/08/2011 02:18 PM, Greg Freemyer wrote:
All,
I just had a btrfs oops with a default 11,4 kernel.
I haven't even tried a update kernel yet since I wasn't sure anyone would care.
Unless advised otherwise, my plan is to try the same test with a Tumbleweed kernel and open a bugzilla if and only if I get the oops there too.
And the award for the most useless bug report goes to... ;)
Seriously, though, btrfs oops are tough to track because error handling is currently implemented by BUG_ON, so the details are important.
- -Jeff
Jeff, As I understand it, the goal is to have btrfs be supported option instead of experimental in 12.1 If so, the Tumbleweed kernel seems to be about as bleeding edge as the factory one and thus a good place to test against. If I'm wasting my time to create a opeSUSE bugzilla let me know and I'll work exclusively with a vanilla kernel and send my reports to kernel.org. fyi: I'm working with a student doing GSoC. And the project is writing tests to test the snapshot features of btrfs (and possibly ext4+snapshots). fyi2: We have ext4+snapshots in a different kernel build, but we've done very little openSUSE specific testing to date. Hopefully I'll have something to talk about related to submitting that to you guys later. Greg -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org
On 07/08/2011 05:25 PM, Greg Freemyer wrote:
On Fri, Jul 8, 2011 at 4:29 PM, Jeff Mahoney<jeffm@suse.de> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 07/08/2011 02:18 PM, Greg Freemyer wrote:
All,
I just had a btrfs oops with a default 11,4 kernel.
I haven't even tried a update kernel yet since I wasn't sure anyone would care.
Unless advised otherwise, my plan is to try the same test with a Tumbleweed kernel and open a bugzilla if and only if I get the oops there too.
And the award for the most useless bug report goes to... ;)
Seriously, though, btrfs oops are tough to track because error handling is currently implemented by BUG_ON, so the details are important.
- -Jeff
Jeff,
As I understand it, the goal is to have btrfs be supported option instead of experimental in 12.1
If so, the Tumbleweed kernel seems to be about as bleeding edge as the factory one and thus a good place to test against.
If I'm wasting my time to create a opeSUSE bugzilla let me know and I'll work exclusively with a vanilla kernel and send my reports to kernel.org.
Nope, you're not wasting your time. We're definitely interested in bringing btrfs up to full speed. It's just that this report was "it crashed" and that's it. If you have an Oops, that'd definitely be helpful. The other issue is that some crashes would normally just be recoverable errors, but the error recovery code hasn't been written yet.
fyi: I'm working with a student doing GSoC. And the project is writing tests to test the snapshot features of btrfs (and possibly ext4+snapshots).
Oh, cool. I bet the btrfs project would be interested in including those with the archive.
fyi2: We have ext4+snapshots in a different kernel build, but we've done very little openSUSE specific testing to date. Hopefully I'll have something to talk about related to submitting that to you guys later.
Cool, I look forward to it. -Jeff -- Jeff Mahoney SuSE Labs -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 07/08/2011 07:29 PM, Jeff Mahoney wrote:
On 07/08/2011 05:25 PM, Greg Freemyer wrote:
On Fri, Jul 8, 2011 at 4:29 PM, Jeff Mahoney<jeffm@suse.de> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 07/08/2011 02:18 PM, Greg Freemyer wrote:
All,
I just had a btrfs oops with a default 11,4 kernel.
I haven't even tried a update kernel yet since I wasn't sure anyone would care.
Unless advised otherwise, my plan is to try the same test with a Tumbleweed kernel and open a bugzilla if and only if I get the oops there too.
And the award for the most useless bug report goes to... ;)
Seriously, though, btrfs oops are tough to track because error handling is currently implemented by BUG_ON, so the details are important.
- -Jeff
Jeff,
As I understand it, the goal is to have btrfs be supported option instead of experimental in 12.1
If so, the Tumbleweed kernel seems to be about as bleeding edge as the factory one and thus a good place to test against.
If I'm wasting my time to create a opeSUSE bugzilla let me know and I'll work exclusively with a vanilla kernel and send my reports to kernel.org.
Nope, you're not wasting your time. We're definitely interested in bringing btrfs up to full speed. It's just that this report was "it crashed" and that's it. If you have an Oops, that'd definitely be helpful. The other issue is that some crashes would normally just be recoverable errors, but the error recovery code hasn't been written yet.
fyi: I'm working with a student doing GSoC. And the project is writing tests to test the snapshot features of btrfs (and possibly ext4+snapshots).
Oh, cool. I bet the btrfs project would be interested in including those with the archive.
fyi2: We have ext4+snapshots in a different kernel build, but we've done very little openSUSE specific testing to date. Hopefully I'll have something to talk about related to submitting that to you guys later.
Cool, I look forward to it.
Just as an aside, in case I sounded sort of flippant WRT btrfs crashes - -- I actually have my personal /data partition (4 TB) running on btrfs with multiple subvolumes. I haven't run into any problems with it. - -Jeff - -- Jeff Mahoney SUSE Labs -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org/ iEYEARECAAYFAk4XrQkACgkQLPWxlyuTD7LlIgCfe1+cPF1SAakhlgNLgX8dJnxN FxEAoIY7r1hHSBqPpA4IuZhDZ7mAjqdH =bKNG -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org
On Fri, Jul 8, 2011 at 7:29 PM, Jeff Mahoney <jeffm@suse.de> wrote:
On 07/08/2011 05:25 PM, Greg Freemyer wrote:
On Fri, Jul 8, 2011 at 4:29 PM, Jeff Mahoney<jeffm@suse.de> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 07/08/2011 02:18 PM, Greg Freemyer wrote:
All,
I just had a btrfs oops with a default 11,4 kernel.
I haven't even tried a update kernel yet since I wasn't sure anyone would care.
Unless advised otherwise, my plan is to try the same test with a Tumbleweed kernel and open a bugzilla if and only if I get the oops there too.
And the award for the most useless bug report goes to... ;)
Seriously, though, btrfs oops are tough to track because error handling is currently implemented by BUG_ON, so the details are important.
- -Jeff
Jeff,
As I understand it, the goal is to have btrfs be supported option instead of experimental in 12.1
If so, the Tumbleweed kernel seems to be about as bleeding edge as the factory one and thus a good place to test against.
If I'm wasting my time to create a opeSUSE bugzilla let me know and I'll work exclusively with a vanilla kernel and send my reports to kernel.org.
Nope, you're not wasting your time. We're definitely interested in bringing btrfs up to full speed. It's just that this report was "it crashed" and that's it. If you have an Oops, that'd definitely be helpful. The other issue is that some crashes would normally just be recoverable errors, but the error recovery code hasn't been written yet.
I opened bugzilla against the 2.6.39 tumbleweed kernel: https://bugzilla.novell.com/show_bug.cgi?id=705039 Hopefully there is enough info there to be useful. Note that this is the first time the oops actually locked my machine. Previously it just caused the test partitions to freeze up. Greg -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org
participants (3)
-
Bruno Friedmann -
Greg Freemyer -
Jeff Mahoney