cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chandra Sekar S (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-12088) Upgrade corrupts SSTables
Date Mon, 19 Sep 2016 16:09:20 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-12088?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15503871#comment-15503871
] 

Chandra Sekar S commented on CASSANDRA-12088:
---------------------------------------------

Sorry, I can't try these anymore. I performed `nodetool scrub` and lost some data as well.
This occurred both in our build and test environment, so it is probably reproducible with
any non-empty upgrade from 2.1 to 3.7. We've decided to dump out data as CSV and import it
into 3.7 for production upgrade.

No, I don't have the old sstables.

> Upgrade corrupts SSTables
> -------------------------
>
>                 Key: CASSANDRA-12088
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-12088
>             Project: Cassandra
>          Issue Type: Bug
>         Environment: OS: CentOS release 6.7 (Final)
> Cassandra version: 2.1, 3.7
>            Reporter: Chandra Sekar S
>            Priority: Critical
>
> When upgrading from 2.0 to 3.7, table was corrupted and an exception occurs when performing
LWT from Java Driver. The server was upgraded from 2.0 to 2.1 and then to 3.7. "nodetool upgradesstables"
was run after each step of upgrade.
> Schema of affected table:
> {code}
> CREATE TABLE payment.tbl (
>     c1 text,
>     c2 timestamp,
>     c3 text,
>     s1 timestamp static,
>     s2 int static,
>     c4 text,
>     PRIMARY KEY (c1, c2)
> ) WITH CLUSTERING ORDER BY (c2 ASC)
>     AND bloom_filter_fp_chance = 0.01
>     AND caching = {'keys': 'ALL', 'rows_per_partition': 'NONE'}
>     AND comment = ''
>     AND compaction = {'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy',
'max_threshold': '32', 'min_threshold': '4'}
>     AND compression = {'chunk_length_in_kb': '64', 'class': 'org.apache.cassandra.io.compress.LZ4Compressor'}
>     AND crc_check_chance = 1.0
>     AND dclocal_read_repair_chance = 0.1
>     AND default_time_to_live = 0
>     AND gc_grace_seconds = 864000
>     AND max_index_interval = 2048
>     AND memtable_flush_period_in_ms = 0
>     AND min_index_interval = 128
>     AND read_repair_chance = 0.0
>     AND speculative_retry = '99PERCENTILE';
> {code}
> Insertion that fails:
> {code:java}
> insert into tbl (c1, s2) values ('value', 0) if not exists;
> {code}
> The stack trace in system.log of cassandra server,
> {code}
> INFO  [HANDSHAKE-maven-repo.corp.zeta.in/10.1.5.13] 2016-06-24 22:23:14,887 OutboundTcpConnection.java:514
- Handshaking version with maven-repo.corp.zeta.in/10.1.5.13
> ERROR [MessagingService-Incoming-/10.1.5.13] 2016-06-24 22:23:14,889 CassandraDaemon.java:217
- Exception in thread Thread[MessagingService-Incoming-/10.1.5.13,5,main]
> java.io.IOError: java.io.IOException: Corrupt flags value for unfiltered partition (isStatic
flag set): 160
> 	at org.apache.cassandra.db.rows.UnfilteredRowIteratorSerializer$1.computeNext(UnfilteredRowIteratorSerializer.java:224)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.db.rows.UnfilteredRowIteratorSerializer$1.computeNext(UnfilteredRowIteratorSerializer.java:212)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.utils.AbstractIterator.hasNext(AbstractIterator.java:47) ~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.db.partitions.PartitionUpdate$PartitionUpdateSerializer.deserialize30(PartitionUpdate.java:681)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.db.partitions.PartitionUpdate$PartitionUpdateSerializer.deserialize(PartitionUpdate.java:642)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.service.paxos.Commit$CommitSerializer.deserialize(Commit.java:131)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.service.paxos.PrepareResponse$PrepareResponseSerializer.deserialize(PrepareResponse.java:97)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.service.paxos.PrepareResponse$PrepareResponseSerializer.deserialize(PrepareResponse.java:66)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.net.MessageIn.read(MessageIn.java:114) ~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.net.IncomingTcpConnection.receiveMessage(IncomingTcpConnection.java:190)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.net.IncomingTcpConnection.receiveMessages(IncomingTcpConnection.java:178)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.net.IncomingTcpConnection.run(IncomingTcpConnection.java:92)
~[apache-cassandra-3.7.0.jar:3.7.0]
> Caused by: java.io.IOException: Corrupt flags value for unfiltered partition (isStatic
flag set): 160
> 	at org.apache.cassandra.db.rows.UnfilteredSerializer.deserialize(UnfilteredSerializer.java:380)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	at org.apache.cassandra.db.rows.UnfilteredRowIteratorSerializer$1.computeNext(UnfilteredRowIteratorSerializer.java:219)
~[apache-cassandra-3.7.0.jar:3.7.0]
> 	... 11 common frames omitted
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message