Checksums #502
Checksums #502
Conversation
Conflicts: src/paxos/master.ml src/paxos/multi_paxos_fsm.ml src/paxos/value.ml src/system/startup.ml src/tlog/tlc2_test.ml src/tlog/tlogcollection_test.ml
|
|
||
| Problem | ||
| ======= | ||
| If a node crashed, and failed to write some tlog entries to disk, this is not detected by Arakoon. The node announces it's in sync up to the last entry in the tlogs, even if other nodes diverged while the node was offline. |
There was a problem hiding this comment.
I think we should mention here that usually it should not be possible for this to happen. Usually means: fsync set to true (which is the default) and none of the layers below (file system, hardware) lie about fsync behaviour.
|
|
||
| let f_entry (i, value) = | ||
| let validate = validate () in | ||
| tlog_coll # log_value_explicit i value ~validate:validate false None >>= fun _ -> |
There was a problem hiding this comment.
You can write this as tlog_coll # log_value_explicit i value ~validate false None >>= fun () ->
note the ~validate and >>= fun ()
| @@ -882,6 +901,8 @@ struct | |||
| >>= fun () -> | |||
| _insert_updates store updates' kt >>= fun (urs:update_result list) -> | |||
| _with_transaction store kt (fun tx -> _incr_i store tx) >>= fun () -> | |||
| let cs = Value.checksum_of value in | |||
| _with_transaction store kt (fun tx -> _set_checksum store cs tx) >>= fun () -> | |||
There was a problem hiding this comment.
I know you're already changing this, but as a reminder to myself: set_checksum should happen in the same transaction as the one that bumps the i just above
|
Should we have some test for the upgrade scenario (as discussed earlier) ? |
| cp -r pylabs/test/* /opt/qbase3/var/tests/arakoon_system_tests/ | ||
| mkdir -p /opt/qbase3/lib/python/site-packages/arakoon/ | ||
| cp src/client/python/*.py /opt/qbase3/lib/python/site-packages/arakoon/ | ||
| sudo mkdir -p /opt/qbase3/apps/arakoon/bin/ |
There was a problem hiding this comment.
These shouldn't be required.
No description provided.