summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* objecter: ping laggy osdsSage Weil2008-10-131-0/+6
| | | | | This ensures we detect any session resets. This necessary because the osd message policy on replies is lossy.
* osd: lossy send policy on acksSage Weil2008-10-131-3/+2
|
* objecter: kick requests when we detect a remote session resetSage Weil2008-10-134-28/+37
|
* msgr: handle incoming lossy connection when outgoing policy is losslessSage Weil2008-10-131-0/+13
| | | | | | Basically, try to wake up our lossless connection. The lossy attempt never survives. This avoids worrying about keeping session state alive (connect_seq, etc.).
* msgr: track lossy mode independently for self, peerSage Weil2008-10-1311-59/+161
| | | | | | | | | The policy will be asymmetrical for the OSDs, so we need to track it independently. The various assertions aren't all worked out yet. Notably, connection races aren't quite right. But the basic bits are there.
* client: fix && || parensSage Weil2008-10-131-1/+1
|
* filestore: use BTRFS_IOC_SYNC instead of fsyncSage Weil2008-10-131-1/+7
| | | | fsync() won't do the right thing on newer versions of btrfs, due do the new tree log stuff. Use the SYNC ioctl.
* filestore: detect new btrfs ioctlsSage Weil2008-10-132-4/+22
| | | Don't use them yet, though.
* fix of crc32c calculationYehuda Sadeh2008-10-131-6/+4
|
* replace the crc32c codeYehuda Sadeh2008-10-114-128/+737
|
* debian: changelog updateSage Weil2008-10-091-1/+6
|
* debian: fix debian-doc; include debian dir in dist tarballSage Weil2008-10-093-4/+2
|
* mon: populate last_beacon properly for mds liveness checkSage Weil2008-10-091-17/+17
| | | | | | It was only populating on paxos updates, but that missed the case where the leader moves from one monitor to another (live) peon, who has an empty last_beacon. Move it where it makes the most sense.
* debian: tweak ceph-doc pathsSage Weil2008-10-092-1/+2
|
* dstart.sh: replace dstartnew.shSage Weil2008-10-093-73/+96
| | | | -n will create new fs, otherwise it will start up using existing data.
* filestore: fix collection attrsSage Weil2008-10-092-22/+33
| | | | Need to prepend user.ceph. to attr names...
* todosSage Weil2008-10-091-5/+9
|
* mon: include pg usage stats, header in pg dumpSage Weil2008-10-091-6/+12
|
* debian: move map tools and crun to ceph-commonSage Weil2008-10-099-12/+17
|
* osd: account for used kb separatelySage Weil2008-10-084-10/+21
| | | | | Account for used separately from available, as the underlying fs may do tricks reserving space for root and such.
* osd: use kb instead of blocks for pg, osd utilizationSage Weil2008-10-086-38/+41
|
* osd: prioritize heartbeat messagesSage Weil2008-10-081-2/+3
|
* msgr: include priority in msg header, make dispatch a priority queueSage Weil2008-10-087-59/+76
| | | | | | | Generalizes previous hack that put messages from the monitor at the front of the dispatch queue. Monitor now just sets a (non-default) default_send_priority of CEPH_MSG_PRIO_HIGH. That value is used only if the message priority isn't set explicitly by set_priority() before being queued for send.
* journal: protect journal access (namely, completions) with mutexSage Weil2008-10-081-0/+9
| | | | The commit_waiters map was getting corrupted occasionally.
* debian: include crun in osd, mds, mon packagesSage Weil2008-10-083-0/+4
|
* osdmaptool: fix silly num_dom bugSage Weil2008-10-081-1/+1
|
* ceph.spec.in: include crunErwin, Brock A2008-10-081-1/+2
|
* kclient: initializing kaddr (merge fix)v0.4Yehuda Sadeh2008-10-072-2/+7
|
* Merge branch 'writepage' into unstableYehuda Sadeh2008-10-076-25/+43
|\ | | | | | | | | | | Conflicts: src/kernel/messenger.c
| * msgr: expand footer flags, include nocrc, kclient nocrc mount optionSage Weil2008-10-076-26/+44
| |
| * kclient: use both writepage and crcYehuda Sadeh2008-10-071-26/+5
| |
| * kclient: disabling the readpageYehuda Sadeh2008-10-071-3/+6
| |
| * kclient: disabling readpageYehuda Sadeh2008-10-071-1/+1
| |
| * kclient: use writepage instead of sendmsg for some of the dataYehuda Sadeh2008-10-071-2/+35
| |
* | kclient: use both writepage and crcYehuda Sadeh2008-10-071-26/+5
| |
* | kclient: disabling the readpageYehuda Sadeh2008-10-071-3/+6
| |
* | kclient: disabling readpageYehuda Sadeh2008-10-071-1/+1
| |
* | kclient: use writepage instead of sendmsg for some of the dataYehuda Sadeh2008-10-071-2/+35
| |
* | kclient: missing include fileYehuda Sadeh2008-10-071-0/+1
| |
* | Revert "kclient: use writepage instead of sendmsg for some of the data"Yehuda Sadeh2008-10-071-35/+2
| | | | | | | | This reverts commit 406fc33a6bb6120925edda0db9fc5f624d43efb9.
* | kclient: use writepage instead of sendmsg for some of the dataYehuda Sadeh2008-10-071-2/+35
| |
* | ebofs: -ENOSPC in apply_transactionSage Weil2008-10-072-0/+16
| |
* | kclient: change default rsize to 128kYehuda Sadeh2008-10-072-3/+3
|/
* kclient: small fix take twoYehuda Sadeh2008-10-071-2/+4
|
* kclient: small fix for bdiYehuda Sadeh2008-10-071-1/+1
|
* kclient: use a private bdiYehuda Sadeh2008-10-073-0/+37
|
* kclient: fix file to object mappingSage Weil2008-10-073-34/+29
| | | | | | We ignore cases where the stripe_unit is small, for now, as that will require passing a stride size back to the caller so that the correctly spaced pages are used.
* crushtool: verify explicit item posSage Weil2008-10-061-0/+4
|
* cm.txt: fix up metaroot poolSage Weil2008-10-061-3/+3
|
* crushtool: verify item posSage Weil2008-10-062-12/+16
|
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-22 01:56:48 +0100