https://bugzilla.redhat.com/show_bug.cgi?id=969327 I'm unable to reproduce this myself, but after much staring at the code and the provided logs (and logs from my own successful runs) I'm worried we may have a problem with hubQ. Notice that in both sets of anaconda logs the "spoke is ready: ..." and "setting <object> status to: ..." come in batches, right next to each other. They are nowhere near the threads that send the messages, most are happening after their associated threads are actually done. This could be a logging artifact, maybe they just aren't being logged until later for some reason. But since the reporters are seeing actual problems I am suspicious that something is holding up hubQ and it is processing them all at about the same time. If you look at the successful run they all happen early. In the failed run they are late, and something in their ordering triggers running AnaExecuteStorageThread a second time which seems to be the source of the actual error. Even in the successful run they don't look the way I'd expect. For example DatetimeSpoke being set ready should be near AnaDateTimeThread, but it isn't -- even in the successful run it is after the thread finishes running. Anyone have any ideas for debugging this? -- Brian C. Lane | Anaconda Team | IRC: bcl #anaconda | Port Orchard, WA (PST8PDT)
Attachment:
pgplZrlKULmoq.pgp
Description: PGP signature
_______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
