On Friday, 13 February 2004 at 21:53:11 +0100, Stijn Hoop wrote: > On Fri, Feb 13, 2004 at 08:15:45PM +0100, Lukas Ertl wrote: >> On Fri, 13 Feb 2004, Stijn Hoop wrote: >> >>> On Fri, Feb 13, 2004 at 05:58:32PM +0100, Lukas Ertl wrote: >>>> On Fri, 13 Feb 2004, Stijn Hoop wrote: >>>>> after I reverted the syscons change I could build a working kernel, but >>>>> something has changed in vinum, so that the volumes are not created in >>>>> /dev/vinum anymore. >>>> >>>> I don't think that a change in vinum caused that (from when was your last >>>> kernel?), but I'm gonna try to reproduce that bug. >>> >>> Last kernel + world were from Feb 6. Unfortunately work got in my way and >>> while my work box has now completed a built from that time, it hasn't been >>> installed yet, and I'll be at home for the weekend so I'd have to install it >>> remotely (which is a bit icky). Maybe I can do a remote install tomorrow, but >>> I wouldn't count on it, so it would be monday at the earliest that I could >>> test other kernels. >> >> Hm, I've now upgraded my testbox to the latest kernel, and I absolutely >> can't reproduce that bug. saveconfig, dumpconfig, starting, stopping, >> everything works fine. :-/ >> >> You're sure the error is not located somewhere else? > > If you could clue me in where to look? The error I see is produced by the > daemon not being able to save the configuration to disk, which as far as my > reading of the source got me. There are a number of things you can do: 1. Give the output of 'vinum list' and 'vinum list -v'. From your ls output, it looks as if it didn't create any objects. Is that correct? If not, does 'saveconfig' work? 2. Check the system log files. It's possible that there's an error message there. > Maybe I can 'dd' the configuration from the disk to send to you? 3. You don't need to do that. Well, probably not. Try 'vinum dumpconfig', which should give you better information. > Maybe it's a failing disk? No, I don't think so. > If so, how could I check? There are no other messages in the syslog. Hmm. Well, that accounts for (2). Greg -- See complete headers for address and phone numbers.
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:43 UTC