Le 22/03/2012 11:59, Sergey B Kirpichev a écrit : > > Well, may be it's ok. For now, I'm not certain that > 1) the severity of this bug IS important (please see this > http://www.debian.org/Bugs/Developer#severities) > 2) patch not break things >
I cannot be sure the patch won't break things. Anyway, it does not seems to because (1) the cause of the problem seems do be clearly identified and (2) the proposed patch is quite old and it does not seems anybody had complained about it. About the severity of the problem, it is definitely not grave or critical because there are many ways to workaround or ignore it, but I feel it important because it leaves thousands of unused values in memory and greatly complicates the analysis of memcached's usage statistics. Let's give up about patching stable if you think it's not necessary. Anyway this "non expiration" can be a great cause of confusion (at least it was for me and I had to go through extensive research to find out what was going on). So, I think some warning about session.gc_maxlifetime not working with this version of the memcached module could be added in /etc/php5/conf.d/memcached.ini -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
