You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@bookkeeper.apache.org by GitBox <gi...@apache.org> on 2022/09/20 04:20:30 UTC

[GitHub] [bookkeeper] zymap commented on issue #3466: Missing ledgers when inducing network packet loss in Bookkeeper 4.15

zymap commented on issue #3466:
URL: https://github.com/apache/bookkeeper/issues/3466#issuecomment-1251816693

   @RaulGracia 
   I saw the logs in step 2:
   
   ```
   2022-07-20T14:45:47.426475848+00:00 stdout F 2022-07-20 14:45:47,426 7065710 [ZKC-connect-executor-0-SendThread(zookeeper-client:2181)] DEBUG org.apache.zookeeper.ClientCnxn - Reading reply session id: 0x2016bc13e0a001c, packet:: clientPath:/pravega/pravega/bookkeeper/ledgers/00/0000/L3246 serverPath:/pravega/pravega/bookkeeper/ledgers/00/0000/L3246 finished:false header:: 146,4  replyHeader:: 146,38654707423,0  request:: '/pravega/pravega/bookkeeper/ledgers/00/0000/L3246,T  response:: #426f6f6b69654d65746164617461466f726d617456657273696f6e933affffff8c18210218020ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff12813238a19626f6f6b6b65657065722d626f6f6b69652d322d3131353830a19626f6f6b6b65657065722d626f6f6b69652d302d32383831361003834204825a16ab6170706c69636174696f6e127507261766567615a15af426f6f6b4b65657065724c6f674964122313360ffffffe2ffffffe5ffffffcdffffff8bffffff81ffffff93ffffffd8ffffff9c10,s{38654707364,38654707364,1658328325290,1658328325290,0,0,0,0,172,0,386
 54707364}
   2022-07-20T14:45:47.426637438+00:00 stdout F 2022-07-20 14:45:47,426 7065710 [BookKeeperClientWorker-OrderedExecutor-7-0] DEBUG o.a.b.client.PendingReadLacOp - Read LAC complete with enough validResponse for ledger: 3246 LAC: -1
   2022-07-20T14:45:47.426737033+00:00 stdout F 2022-07-20 14:45:47,426 7065710 [core-13] DEBUG o.a.b.meta.AbstractZkLedgerManager - Unregistered ledger metadata listener ReadOnlyLedgerHandle(lid = 3246, id = 672259987) on ledger 3246.
   2022-07-20T14:46:16.612224510+00:00 stdout F 2022-07-20 14:46:16,612 7094896 [BookKeeperClientWorker-OrderedExecutor-7-0] DEBUG o.a.b.proto.PerChannelBookieClient - Could not write Add request to bookie 192.169.1.60/192.169.1.60:3181 for ledger 3246, entry 0
   2022-07-20T14:46:16.612271701+00:00 stdout F 2022-07-20 14:46:16,612 7094896 [BookKeeperClientWorker-OrderedExecutor-7-0] ERROR o.a.bookkeeper.client.PendingAddOp - Write of ledger entry to quorum failed: L3246 E0
   2022-07-20T14:46:16.616440402+00:00 stdout F 2022-07-20 14:46:16,616 7094900 [BookKeeperClientWorker-OrderedExecutor-7-0] DEBUG o.a.b.proto.PerChannelBookieClient - Could not write ReadLAC request to bookie null for ledger 3246, entry -1
   ```
   
   There have two things I noticed:
   
   1. The ledger seems closed after it received the ReadLAC response.
   
   ```
   Read LAC complete with enough validResponse for ledger: 3246 LAC: -1
   2022-07-20T14:45:47.426737033+00:00 stdout F 2022-07-20 14:45:47,426 7065710 [core-13] DEBUG o.a.b.meta.AbstractZkLedgerManager - Unregistered ledger metadata listener ReadOnlyLedgerHandle(lid = 3246, id = 672259987) on ledger 3246
   ```
   
   2. Bookie channel is null in the callback. That looks like when the callback is triggered, the channel is close.
   
   ```
   Could not write ReadLAC request to bookie null for ledger 3246, entry -1
   ```
   
   And looks like in the Pravega, the ledger handler closes again so the error happens again. (step 2 closed and step 3 closed again)
   
   ```
   2022-07-20T14:46:16.630253410+00:00 stdout F 2022-07-20 14:46:16,629 7094913 [core-30] DEBUG o.a.bookkeeper.client.LedgerHandle - Failed to close ledger 3246 :
   2022-07-20T14:46:16.630795088+00:00 stdout F 2022-07-20 14:46:16,630 7094914 [core-30] ERROR i.p.s.s.i.bookkeeper.BookKeeperLog - Log[13]: Unable to close LedgerHandle for Ledger 3246.
   2022-07-20T14:46:16.630795088+00:00 stdout F io.pravega.segmentstore.storage.DurableDataLogException: Unable to close ledger 3246.
   ```
   
   -----
   
   I saw in the previous, you are using 4.14.1, I found there has [a change](https://github.com/apache/bookkeeper/pull/2794/files) related to this which was released in 4.14.3. 
   
   Could you please also run the test against with the 4.14.1 and 4.14.3?


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@bookkeeper.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org