On Sun, 2012-03-11 at 18:20 +0100, Harald Dunkel wrote: > On 03/09/12 15:57, Ben Hutchings wrote: > > On Fri, 2012-03-09 at 13:30 +0100, Harald Dunkel wrote: > >> PS: I just noticed that severity is set to "normal". Sorry > >> to say, but I disagree on the severity in this case. If our > >> production environment dies after 200 days uptime, then this > >> is fatal. > > > > Why do you say '200 days uptime'? > > > > The division by zero came up on several servers in my environment > after more than 200 days uptime each. I have never seen this bug > pop up immediately. Looking at > > https://bugzilla.kernel.org/show_bug.cgi?id=16991 > > it seems that an uptime of several months before being hit by the > problem is not unusual. > > (Novell had a 200 days uptime problem with their 2.6.32 kernel, too, > even though I am not sure that this is the same problem: > > http://www.novell.com/support/viewContent.do?externalId=7009834&sliceId=1 > ) > > Anyway, does the uptime matter? A crashing server in a production > environment is a severe problem, regardless how long the machine > was up before.
There was a bug that caused systems to crash after 208 days, which the Novell page refers to. That was fixed in longterm update 2.6.32.50 and Debian's version 2.6.32-40. But other people report this crash occurring after a much shorter uptime: https://bugzilla.kernel.org/show_bug.cgi?id=16991#c12 https://bugzilla.kernel.org/show_bug.cgi?id=16991#c27 https://bugzilla.kernel.org/show_bug.cgi?id=16991#c28 So I would say there is more than one bug that can cause these assertions to fail. > >> Would you mind to adjust the severity of this bug report? > > > > We have what is supposed to be a workaround. Does it not work? Have > > you seen any warnings? > > > > In which Debian kernel can I find the workaround? 2.6.32-36 Ben. -- Ben Hutchings For every action, there is an equal and opposite criticism. - Harrison
signature.asc
Description: This is a digitally signed message part
