Alfresco reindexer issue!

[rajd]

Hello all,

Recently our Alfresco install crashed, which corrupted the Lucene indexes.
So I configured Alfresco (in repository.properties) to do a FULL Lucene reindex on the next Alfresco start.

Alfresco started just fine and started reindexing as expeced. However, after more than 24 hours, the indexer still is running.

The first few hours the server was under high load (15 and more) and this suddenly dropped down to exactly 1. With that drop came a drop of I/O activity as well.

From the logs:

01:47:01,674 INFO  [org.alfresco.repo.node.index.FullIndexRecoveryComponent]    10 % complete.
04:08:22,419 INFO  [org.alfresco.repo.node.index.FullIndexRecoveryComponent]    20 % complete.
04:52:33,774 ERROR [org.alfresco.repo.node.index.AbstractReindexComponent.threads] Reindexer   302 failed with error: 10180000 Exception in Transaction..
04:53:38,857 ERROR [org.alfresco.repo.node.index.AbstractReindexComponent.threads] Reindexer   303 failed with error: 10180001 Exception in Transaction..
05:33:03,788 ERROR [org.alfresco.repo.node.index.AbstractReindexComponent.threads] Reindexer   312 failed with error: 10180004 Exception in Transaction..
05:33:04,010 ERROR [org.alfresco.repo.node.index.AbstractReindexComponent.threads] Reindexer   315 failed with error: 10180003 Failed to open stream onto channel: 
   accessor: ContentAccessor[ contentUrl=store://2009/7/29/16/19/442f8a9b-933d-47df-b77f-cb65c7c3b20d.bin, mimetype=application/pdf, size=41927, encoding=UTF-8, locale=en_U
S].
05:33:04,409 ERROR [org.alfresco.repo.node.index.AbstractReindexComponent.threads] Reindexer   319 failed with error: 10180005 Exception in Transaction..
07:32:01,241 ERROR [org.alfresco.repo.node.index.AbstractReindexComponent.threads] Reindexer   350 failed with error: 10180006 Exception in Transaction..
09:10:16,691 ERROR [org.alfresco.repo.node.index.AbstractReindexComponent.threads] Reindexer   355 failed with error: 10180007 Exception in Transaction..
‍‍‍‍‍‍‍‍‍‍‍‍‍

Now I'm wondering: is the indexer still running (the Linux process lists suggest so, as Java is constantly consuming 1 CPU core) and is this behaviour normal?

Some info:
Alfresco 3.2 CE
CentOS 5.4 64 bit server, 16GB of memory, 4 CPU cores
JDK 1.6, Tomat 6.0.26
About 250000 PDF files in Alfresco

Thanks!

[fracat71]

Hi,
i have the same problem. Someone can help us?
This is my situation: a folder tree with 500000 folder without documents.

17:46:38,968  ERROR [index.AbstractReindexComponent.threads] Reindexer    12 failed with error: 00030009 Failed to prepare: requires rollback.
17:46:39,122  ERROR [index.AbstractReindexComponent.threads] Reindexer    13 failed with error: 00030005 Failed to prepare: requires rollback.
17:46:39,186  ERROR [index.AbstractReindexComponent.threads] Reindexer    14 failed with error: 00030003 Failed to prepare: requires rollback.
17:46:39,263  ERROR [index.AbstractReindexComponent.threads] Reindexer    15 failed with error: 00030007 Failed to prepare: requires rollback.
18:13:22,357  INFO  [node.index.FullIndexRecoveryComponent]  10 % complete. 
18:43:59,711  INFO  [node.index.FullIndexRecoveryComponent]  20 % complete. 
19:14:26,183  INFO  [node.index.FullIndexRecoveryComponent]  30 % complete. 
19:43:27,238  INFO  [node.index.FullIndexRecoveryComponent]  40 % complete. 
20:14:08,026  INFO  [node.index.FullIndexRecoveryComponent]  50 % complete. 
20:43:13,312  INFO  [node.index.FullIndexRecoveryComponent]  60 % complete. 
21:10:57,492  INFO  [node.index.FullIndexRecoveryComponent]  70 % complete. 
21:39:40,309  INFO  [node.index.FullIndexRecoveryComponent]  80 % complete. 
22:08:08,288  INFO  [node.index.FullIndexRecoveryComponent]  90 % complete. 
22:41:14,791  INFO  [node.index.FullIndexRecoveryComponent]  100 % complete.‍‍‍‍‍‍‍‍‍‍‍‍‍‍

Thanks

[muralie39]

create new instance & use auto mode & place the old dir.root total folders in the new dor.root path