Bug 837941
Summary: | Data loss when writing to qcow2-format disk files | ||
---|---|---|---|
Product: | Red Hat Enterprise Linux 7 | Reporter: | Richard W.M. Jones <rjones> |
Component: | libguestfs | Assignee: | Richard W.M. Jones <rjones> |
Status: | CLOSED CURRENTRELEASE | QA Contact: | Virtualization Bugs <virt-bugs> |
Severity: | unspecified | Docs Contact: | |
Priority: | unspecified | ||
Version: | 7.0 | CC: | acathrow, bfan, juzhang, knoel, leiwang, mbooth, qguan, rjones |
Target Milestone: | rc | ||
Target Release: | --- | ||
Hardware: | Unspecified | ||
OS: | Unspecified | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | 837691 | Environment: | |
Last Closed: | 2013-07-18 16:01:08 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | 836710 | ||
Bug Blocks: |
Description
Richard W.M. Jones
2012-07-05 21:18:51 UTC
Hello, I tried your script and do more than 500 runs, it has no error output. https://bugzilla.redhat.com/show_bug.cgi?id=836710#c5 Is there need other steps to repoduce? (In reply to bfan from comment #2) > Hello, > I tried your script and do more than 500 runs, it has no error output. > https://bugzilla.redhat.com/show_bug.cgi?id=836710#c5 > > Is there need other steps to repoduce? This was actually a bug in qemu. The problem was that qemu's qcow2 code used to crash, and libguestfs didn't detect the crash (so instead you would just get occasional lost data writes). Since then: - qemu was fixed so the qcow2 code no longer crashes - the kernel was fixed so it issues a write barrier for sync - libguestfs has been modified to add a guestfs_shutdown call to detect these crashes and report the error instead of ignoring it. Since this was all fixed a long time ago, it's unlikely that you will be able to reproduce the bug. You would have to go back to some quite old versions of qemu, kernel and libguestfs. My suggestion is that we close this bug. Even if the qemu bug reappears, libguestfs will now detect and report it. I am closing this, based on my previous comment. |