Michael Meskes wrote: >> I'm not saying, that quota is at fault here. I reassigned the bug mostly > > So why did you reassign the bug to quota?
I don't want to play BTS ping-pong, but I explained why. Please reread my emails. > >> because it's imho not a bug in network-manager. network-manager is >> simply not conceived for such a use case. > > Other users with similar problems might find the report if it's listed > with network-manager but they certainly won't find it under quota. I don't understand this logic. This error only happens when you have a setup which uses quota and it's the quota process that hangs. So, the first thing people will be looking for (as did John), is a problem with quota. >> network-manager is intended for highly dynamic network environments, >> where networks come and go and where you switch between different networks. >> Removing network-manager is probably the best "bug fix" there is atm. >> For a static configuration with NFS mounts etc, as Johns' setup, I would >> recommend to use plain ifupdown. > > If this is the only solution, fine, but it has to be documented IMO. Who said it's the only solution? >> P.S.: What happens if you pull the network plug? Will quota then also >> fail to stop? Is this a matter of how the nfs share was mounted (hard or >> soft)? > > Problem is (at least I think it is) that quota stats the mountpoint which > then fails because the server does not respond anymore (obviously). > > But I don't think it makes sense to discuss this detail because more > severe problems seem to arise from this setup. Well, then it probably it should be documented within the quota package, that it can't deal with network outages. Cheers, Michael -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth?
signature.asc
Description: OpenPGP digital signature
