This bug was fixed in the package memcached - 1.4.2-1ubuntu4
---
memcached (1.4.2-1ubuntu4) lucid-proposed; urgency=low
* debian/patches/fix-issue-102-segfault.patch: use strncmp
the way the upstream code does. Prevents multigets from
failing. (LP: #637114)
* debian/patche
Hi
With the package in -proposed I got consistent results when executing
the test script for reproducing the issue:
array(0) {
}
array(0) {
}
int(0)
I was able to confirm the issue prior to testing as well.
Thanks.
** Tags removed: verification-needed
** Tags added: verification-done
--
You
** Branch linked: lp:ubuntu/lucid-proposed/memcached
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/637114
Title:
Large multiget requests randomly broken
To manage notifications about this bug go to
Hello Carlos, or anyone else affected,
Accepted memcached into lucid-proposed, the package will build now and
be available in a few hours. Please test and give feedback here. See
https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to
enable and use -proposed. Thank you in advance!
Uploaded fix to lucid-proposed.
** Changed in: memcached (Ubuntu Lucid)
Status: Triaged => Fix Committed
** Description changed:
+ == SRU REPORT ==
+
+ IMPACT: users who try to use memcached with large multi-gets will get a
+ cache miss every time.
+
+ TEST CASE:
+
+ * install php5-cli
The cause of this is a broken back port of the upstream change set:
https://github.com/memcached/memcached/commit/d9cd01ede97f4145af9781d448c62a3318952719
That changeset uses strncmp whereas the .diff file uses strcmp
++if (ptr - c->rcurr > 100 ||
++ strcmp(
Carlos, thanks for taking the time to file this bug report.
I was able to confirm this on lucid only. It is working fine in Maverick
and Natty.
Since its hard to know that this is affecting you, and memcached can
potentially be used as a sort of data store, I'm marking it as High
Importance.
**
