Comment #5 on issue 370 by [email protected]: memcached(v1.4.15) dead loop
bug.
http://code.google.com/p/memcached/issues/detail?id=370
You should be upgrading occasionally; putting 1-2 servers on a new version,
and slowly rolling it out. Otherwise we end up in this sort of a mexican
standoff. 1.4.15 is from 2012.
The problem is that the bug we fixed causes essentially random memory
corruption. As I noted in the PR I was able to, by manipulating when and
where the structures leaked manually with gdb, double link items into the
hash table. It'd be possible to get them to link to themselves.
Every other person who's reported a bug like this, I've had them presize
the hash table as one of the troubleshooting steps (-o hashpower=n) to
avoid hash expansion, and it still ends up happening. I really wanted to
blame that code (and it might still be problematic!) but everything ended
up looking related to this other bug we fixed in .19.
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
--
---
You received this message because you are subscribed to the Google Groups "memcached" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
