Log: http://conference.openafs.org/[email protected]/2014-02-12.txt
Participants: * Andrew Deason * Ben Kaduk * Derrick Brashear * Jason Edgecombe * Ken Dreyer * Marc Dionne * Mike Meffie * Stephan Wiesand == Linux news == Marc tested with the current mainline, no need for changes yet. == 1.6.7 changes pre announcements == Last week, we decided to make 1.6.7+ clients give up their callbacks on shutdown, causing problems for older fileservers in the 1.3.50 - 1.4.5 and 1.5.0 - 1.5.27 ranges, without a knob to turn this off, and that this should be announced in advance on the -info and -announce lists as well as at EAKC. It just wasn't defined who will do it, and when. Postponed to a meeting with Jeffrey present, since it was his proposal. Stephan proposes including the planned fileserver change to -sync=onclose in these announcements. == Problem reports == Linux mountpoint dentries See https://lists.openafs.org/pipermail/openafs-info/2014-February/040503.html Andrew's summary: It was seen via discussion (with Jason/Stephan) that the 130273 issue is relevant for new RHEL5 updates, as well as RHEL6. RHEL7 is not affected yet, but that could change. Andrew is to get a little more information and reproduce the relevant scenarios and we'll talk about this next week. Andrew may try to bring the "bind-mount" implementation up to speed if he has time. == Status of other items foreseen for 1.6.7 == Marc's patch for the Linux getcwd ENOENT problem (see RT #131780 and references therein), gerrit 10804, was merged during the meeting. As well as 10598 which hadn't quite made 1.6.6. And a few others were merged or reviewed sufficiently to be merged once their dependencies are. Stephan complains about there being too many path conflicts between changes pulled up for 1.6 in gerrit already, and asks to not add more. [the remainder of this text is 1.6.7 bookeeping] Idledead change to help the "no buffer space available" problem: Gerrit 10773 ist still being reviewed on master, blocked on two trivial changes to be done. Parallel build fixes (http://gerrit.openafs.org/#q,project:openafs+branch:master+topic:parallel-make,n,z ). No progress. === Accepted on master, to be pulled up to 1_6_x === - nothing left that was discussed before - === Pulled up to 1_6_x, to be reviewed === Ben's changes to allow using the OS X buildbot slave for 1_6_x too: 10742 afs_fetchstore: avoid use of uninitialized variable (still carries a -1, addition in 10835) 10744 afs: Handle reading past the end of a file (blocked on solving a problem it might mask on master first) 10745 libafs: Remove afs_read duplication (needs review) 10835 afs_fetchstore: re-avoid uninitialized variable (needs review) Interrupt clients accessing offlining volumes: 10799 viced: Add options for interrupting clients 6266 vol: Interrupt RX calls accessing offlining vols Fileserver sync behavior: 10809 ihandle: Make _ONCLOSE the sync behavior default Give up callbacks at shutdown: 8840 libafs: Correct GUACB afs_LoopServers flags 6272 unix: giveupallcallbacks at shutdown Others (new pullups): 10813 afs: Treat vc_error as a CheckCode-translated code 10811 salvager: Handle multiple/inconsistent linktables 10810 namei: Ignore misplaced files Others (no progress, or insufficient progress to merge): 10772 vol: Nuke parent vol special inodes 10771 salvager: Ignore linktable-only RW volumes 10770 namei: Set inconsistent linktable linkCount to 0 10769 salvager: Fix in-memory invalid linktable counts 10759 viced: Avoid issuing redundant TMAY requests 10758 viced: Move host quota calculation 10757 viced: Improve client error log messages 10756 viced: Improve CallPreamble error messages 10755 viced: Avoid endless BCB loop 10603 cellconfig: Do not use 'long' for dbserver IPs More changes were pulled up to 1.6.x recently, without prior discussion and adding more path conflicts. These are ignored here and marked -2 in gerrit for the time being. -- Stephan Wiesand DESY -DV- Platanenenallee 6 15738 Zeuthen, Germany _______________________________________________ OpenAFS-devel mailing list [email protected] https://lists.openafs.org/mailman/listinfo/openafs-devel
