- 14 Oct, 2010 40 commits
-
-
Lars Ellenberg authored
Consolidate the ifdef's for the debug level, accidentally the used both DEBUG and DRBD_DEBUG_MD_SYNC. Default to off. For production, we can safely reduce the grace period for this timer again the the value we used to have. Signed-off-by:
Philipp Reisner <philipp.reisner@linbit.com> Signed-off-by:
Lars Ellenberg <lars.ellenberg@linbit.com>
-
Lars Ellenberg authored
It sometimes may take a while for the after state change work to be scheduled, which does drbd_md_sync. At convenient places, we should do explicit drbd_md_sync to have the new state information on disk as soon as possible. Signed-off-by:
-
Lars Ellenberg authored
commit 2372c38caadeaebc68a5ee190782c2a0df01edc3 drbd: fix for possible deadlock on IO error during resync introduced a new ASSERT, which turns out to be wrong. Drop it. Also serialize the state change to D_DISKLESS with the after state change work of the -> D_FAILED transition, don't open a new race. Signed-off-by:
-
Lars Ellenberg authored
Signed-off-by:
-
Lars Ellenberg authored
As we usually update the generation UUIDs here, we should explicitly sync them to disk. So far this has been done only implicitly by related code paths. Signed-off-by:
-
Philipp Reisner authored
This might happen if on the VERIFY_S node the disk gets dropped. Although this is an cluster wide state transition, the VERIFY_T node, updates it connection state first. Then the ack packet for the cluster wide state transition travels back, and the VERIFY_S node stops to produce the P_OV_REQUEST packets. There is absolutely nothing wrong with that. Further, do not log "Can not satisfy peer's..." on the VERIFY_S node in this case, but pretend that they had equal checksum. [Bugz 327] Signed-off-by:
-
Lars Ellenberg authored
Scenario: Something (say, flush-147:0) is in drbd_al_begin_io, holding a local_cnt, waiting for the resync to make progress. Disk fails, worker in after_state_ch does drbd_rs_cancel_all, then waits for local_cnt to drop to zero. flush-147:0 is woken by drbd_rs_cancel_all, needs to write an AL transaction, and queues that on the worker. Deadlock. Fix: do not wait in the worker, have put_ldev() trigger the state change D_FAILED -> D_DISKLESS when necessary. put_ldev() cannot do the state change directly, as it may or may not already hold various spinlocks. We queue a short work instead. Signed-off-by:
-
Lars Ellenberg authored
Various cleanup paths have been incomplete, for the very unlikely case that we cannot allocate enough bios from process context when submitting on behalf of the peer or resync process. Never observed. Signed-off-by:
-
Lars Ellenberg authored
If it was an "empty" resync, the SyncSource may have already "finished" the resync and rotated the UUIDs, before noticing the connection loss (and generating a new uuid, if Primary, rotating again), while the SyncTarget did not change its uuids at all, or only got to the previous sync-uuid. This would then again lead to a full sync on next handshake (see also Bug #251). Fix: Use explicit resync finished notification even for empty resyncs, do not finish an empty resync implicitly on the SyncSource. Signed-off-by:
-
Lars Ellenberg authored
Preparation patch so more drbd_send_state() usage on the peer will not confuse drbd in receive_state(). Signed-off-by:
-
Lars Ellenberg authored
no functional change, just using full state instead of just the .conn part of it for comparisons. Signed-off-by:
-
Lars Ellenberg authored
drbd commit 17c854fea474a5eb3cfa12e4fb019e46debbc4ec drbd: receiving of big packets, for payloads between 64kByte and 4GByte introduced a new on-the-wire packet header format. We must no longer assume either format, but use the result of whatever drbd_recv_header has decoded. Signed-off-by:
-
Lars Ellenberg authored
We used to be16_to_cpu the length field in our received packet header. drbd commit 17c854fea474a5eb3cfa12e4fb019e46debbc4ec drbd: receiving of big packets, for payloads between 64kByte and 4GByte changed this, but forgot to adjust a few places where we relied on h->length being in native byte order. This broke the receiving side of the RLE compressed bitmap exchange. Signed-off-by: -
Philipp Reisner authored
This caused rs_planed to be not in sync with the content of the fifo. That in turn could cause that the resync comes to a complete halt. Signed-off-by:
-
Philipp Reisner authored
Connections through a compressing proxy might have more bits on the fly. 500MByte instead of 50MByte Signed-off-by:
-
Lars Ellenberg authored
If we release the page pointed to by md_io_tmpp, we need to zero out the pointer, too, as that may be used later to decide whether we need to allocate a new page again. Impact: a previously freed page may be used and clobbered. Depending on what that particular page is being used for meanwhile, this may result in silent data corruption of completely unrelated things. Only of concern on devices with logical_block_size != 512 byte, if you re-attach after becoming diskless once. Signed-off-by:
-
Lars Ellenberg authored
Two missing corner cases to the "maximum packet size" handshake. Signed-off-by:
-
Philipp Reisner authored
There are three ways to get IO suspended: * Loss of any access to data * Fence-peer-handler running * User requested to suspend IO Track those in different bits, so that one condition clearing its state bit does not interfere with the other two conditions. Only when the user resumes IO he overrules all three bits. The fact is hidden from the user, he sees only a single suspend bit. Signed-off-by:
-
Lars Ellenberg authored
Signed-off-by:
-
Philipp Reisner authored
Forgot to consider the max size for the resync requests. Signed-off-by:
-
Lars Ellenberg authored
If a synctarget lost connection while being WFSyncUUID, due to "state sanitizing", the attempted state change to SyncTarget looked like an "invalidate" to after_state_ch() later, thus caused a full sync on next handshake (Bug #318). drbd0: PingAck did not arrive in time. drbd0: peer( Primary -> Unknown ) conn( WFSyncUUID -> NetworkFailure ) pdsk( UpToDate -> DUnknown ) from : { cs:NetworkFailure ro:Secondary/Unknown ds:UpToDate/DUnknown r--- } to : { cs:SyncTarget ro:Secondary/Unknown ds:Inconsistent/DUnknown r--- } after sanizising, resulted in state: { cs:NetworkFailure ro:Secondary/Unknown ds:Inconsistent/DUnknown r--- } drbd0: disk( UpToDate -> Inconsistent ) Fix: don't mask state transition errors in "sanitizing", so the requested state change to SyncTarget fails, instead of being implicitly "remaped" to invalidate. Signed-off-by: -
Lars Ellenberg authored
Signed-off-by:
-
Lars Ellenberg authored
If we cannot satisfy a request (because our disk just broke), we still need to drain the payload. Or we'll get a protocol error when interpreting the payload as DRBD packet header. Signed-off-by:
-
Lars Ellenberg authored
BUG trace would look like: lc_find drbd_rs_complete_io got_OVResult drbd_asender Could be triggered by explicit, or IO-error policy based, detach during online-verify. We may only dereference mdev->resync, if we first get_ldev(), as the disk may break any time, causing mdev->resync to disappear once all ldev references have been returned. Already in flight online-verify requests or replies may still come in, which we then need to ignore. Signed-off-by:
-
Lars Ellenberg authored
Signed-off-by:
-
Philipp Reisner authored
Signed-off-by:
-
Lars Ellenberg authored
Just in case we have some pending meta data changes to sync, do it before we call our userland helper, as that may take some time, or even cause a hard reboot. Signed-off-by:
-
Lars Ellenberg authored
addendum to baa33ae4eaa4477b60af7c434c0ddd1d182c1ae7 The race: drbd_md_sync() if (!test_and_clear_bit(MD_DIRTY, &mdev->flags)) return; ==> RACE with drbd_md_mark_dirty() rearming the timer. del_timer(&mdev->md_sync_timer); Fixed by moving the del_timer before the test_and_clear_bit. Additionally only rearm the timer in drbd_md_mark_dirty, if MD_DIRTY was not already set, reduce the grace period from five to one second, and add an ifdef'ed debuging aid to find code paths missing an explicit drbd_md_sync, if any, as those are the only relevant ones for this race. Signed-off-by: -
Philipp Reisner authored
The actual race happened int the drbd_start_resync() function. Where drbd_resync_finished() -> __drbd_set_state() set STOP_SYNC_TIMER and armed the timer. If the timer fired before execution reaches the mod_timer statement at the end of drbd_start_resync() the latter would cause an unexpected call to w_make_resync_request(). Removed the STOP_SYNC_TIMER bit, and base it on the connection state. The STOP_SYNC_TIMER bit probably originates probably the time before the state engine. Signed-off-by:
-
Lars Ellenberg authored
If pacemaker (for example) decided to initialize minor devices not in the exact sync-after dependency order, the configuration partially failed with an error "The sync-after minor number is invalid". (Bugz. #322) We can avoid that by implicitly creating unconfigured minor devices, if others depend on them. Signed-off-by:
-
Lars Ellenberg authored
The race: drbd_md_mark_dirty() drbd_md_sync() if (!test_and_clear_bit(MD_DIRTY, &mdev->flags)) return; drbd_md_sync_page_io(mdev, mdev->ldev, sector, WRITE) ==> RACE clear_bit(MD_DIRTY, &mdev->flags); <== spurious Fixed by removing the spurious clear_bit. Signed-off-by:
-
Lars Ellenberg authored
If a drbd_nl_net_conf hits the small window between the state change to C_STANDALONE and the corresponding cleanup in after_state_ch, that cleanup would throw away stuff we now need again, and later trigger BUG_ON()s. Fixed by properly serializing the new config request with any pending cleanup. Signed-off-by:
-
Philipp Reisner authored
When the complete device is marked as out of sync, we can disable updates of the on disk AL. Currently AL updates are only disabled if one uses the "invalidate-remote" command on an unconnected, primary device, or when at attach time all bits in the bitmap are set. As of now, AL updated do not get disabled when a all bits becomes set due to application writes to an unconnected DRBD device. While this is a missing feature, it is not considered important, and might get added later. BTW, after initializing a "one legged" DRBD device drbdadm create-md resX drbdadm -- --force primary resX AL updates also get disabled, until the first connect. Signed-off-by:
-
Philipp Reisner authored
Now we have multiple BIOs per ee, packets with a 32 bit length field, it gets time to use these goodies. Signed-off-by:
-
Philipp Reisner authored
Signed-off-by:
-
Philipp Reisner authored
Signed-off-by:
-
Philipp Reisner authored
If we intent to use the block_id member of an epoch entry, we may not use the digest member. Signed-off-by:
-
Philipp Reisner authored
Signed-off-by:
-
Lars Ellenberg authored
We now track the data rate of locally submitted resync related requests, and can thus detect non-resync activity on the lower level device. If the current sync rate is above c-min-rate, and the lower level device appears to be busy, we throttle the resyncer. Signed-off-by:
-
Lars Ellenberg authored
also canonicalize the return values of read_for_csum and drbd_rs_begin_io to return -ESOMETHING, or 0 for success. Signed-off-by:
-
