diff options
| author | Lennart Poettering <lennart@poettering.net> | 2014-12-30 20:57:53 +0100 |
|---|---|---|
| committer | Lennart Poettering <lennart@poettering.net> | 2015-01-05 01:40:51 +0100 |
| commit | fa6ac76083b8ffc1309876459f54f9f0e2843731 (patch) | |
| tree | 5169233d54825b07e63da5d11fbc3b4e248aa38b /src/journal/journald-server.c | |
| parent | f93bf4363395018ef48d744c4624158623afd693 (diff) | |
journald: process SIGBUS for the memory maps we set up
Even though we use fallocate() it appears that file systems like btrfs
will trigger SIGBUS on certain low-disk-space situation. We should
handle that, hence catch the signal, add it to a list of invalidated
pages, and replace the page with an empty memory area. After each write
check if SIGBUS was triggered, and consider the write invalid if it was.
This should make journald a lot more robust with file systems where
fallocate() is not reliable, for example all CoW file systems
(btrfs...), where changing written data can fail with disk full errors.
https://bugzilla.redhat.com/show_bug.cgi?id=1045810
Diffstat (limited to 'src/journal/journald-server.c')
| -rw-r--r-- | src/journal/journald-server.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/src/journal/journald-server.c b/src/journal/journald-server.c index a2a2e197c0..6d037cfec4 100644 --- a/src/journal/journald-server.c +++ b/src/journal/journald-server.c @@ -452,6 +452,7 @@ bool shall_try_append_again(JournalFile *f, int r) { -EFBIG Hit fs limit -EDQUOT Quota limit hit -ENOSPC Disk full + -EIO I/O error of some kind (mmap) -EHOSTDOWN Other machine -EBUSY Unclean shutdown -EPROTONOSUPPORT Unsupported feature @@ -469,6 +470,8 @@ bool shall_try_append_again(JournalFile *f, int r) { log_info("%s: Unsupported feature, rotating.", f->path); else if (r == -EBADMSG || r == -ENODATA || r == ESHUTDOWN) log_warning("%s: Journal file corrupted, rotating.", f->path); + else if (r == -EIO) + log_warning("%s: IO error, rotating.", f->path); else return false; |