Vertica: Copy cluster Fail version 7.2.1
I have 2 cluster and 3 node per cluster.
I try to set up copy cluster and run command
/opt/vertica/bin/vbr.py -t copycluster --config-file /home/dbadmin/backup_script/copy_cluster.ini
found message:
Configured backup directories in Mapping are ignored for copycluster.
Starting copy of database xxxxx_DW.
Participating nodes: v_xxxxx_dw_node0001, v_xxxxx_dw_node0002, v_xxxxx_dw_node0003.
Snapshotting database.
Snapshot complete.
Syncing data to destination cluster.
[==================================================] 100%
Reinitializing destination catalog.
Error: Catalog bootstrap failed.
Copycluster FAILED.
how to resolve this problem? How can I solve it?
0
Comments
Hi,
Can you share me the .ini file used for copy cluster and also the vertica.log located in catalog directory of the source cluster.
Thank you,
Sruthi
hi,
Thanks for your help . Here is the config file and log
snapshotName = full_backup
restorePointLimit = 4
tempDir = /tmp/vbr_rsyncd
retryCount = 2
retryDelay = 1
[Database]
dbName = xxxxx_DW
dbUser = dbadmin
dbPassword = dbadmin1
dbPromptForPassword = False
[Transmission]
encrypt = False
checksum = False
port_rsync = 50000
[Mapping]
v_xxxxx_dw_node0001= bk-vtgpoc-1101:/backup
v_xxxxx_dw_node0002= bk-vtgpoc-1102:/backup
v_xxxxx_dw_node0003= bk-vtgpoc-1103:/backup
Here is the log
thanks
Hi,
Vertica.log shared to me contains log from 2016-02-16 22:20:07.077.I could not find any instance of copycluster in the log. Did you run copylcuster before that time? If so could please run it now and share me the log. Please map the node names with IP addresses instead of host names in .ini file and then give a try. Are source and target cluster on 7.2.1??
2016-02-16 22:20:07.077 INFO New log
2016-02-16 22:20:07.077 unknown:0x7f0ce100a720 [Init] <INFO> Log /agoda/Agoda_DW/Agoda_DW/v_agoda_dw_node0001_catalog/vertica.log opened; #1
2016-02-16 22:20:07.077 unknown:0x7f0ce100a720 [Init] <INFO> Processing command line: /opt/vertica/bin/vertica -D /agoda/Agoda_DW/Agoda_DW/v_agoda_dw_node0001_catalog -C Agoda_DW -n v_agoda_dw_node0001 -h 192.168.6.1 -p 5433 -P 4803 -Y ipv4
2016-02-16 22:20:07.077 unknown:0x7f0ce100a720 [Init] <INFO> Starting up Vertica Analytic Database v7.2.1-0
2016-02-16 22:20:07.077 unknown:0x7f0ce100a720 [Init] <INFO> Project Codename: Excavator
2016-02-16 22:20:07.077 unknown:0x7f0ce100a720 [Init] <INFO> vertica(v7.2.1-0) built by jeeves@build-centos6 from releases/VER_7_2_RELEASE_BUILD_1_0_20151116@177918 on 'Mon Nov 16 16:59:07 2015' $BuildId$
2016-02-16 22:20:07.077 unkno
hi
Thanks for your suggestion. I try to change the .ini file using io-address instead of host name and rerun copy cluster process again. source and dest cluster is version 7.2.1
It shows the message:
[dbadmin@bk-vtgpoc-1001 agoda]$ /opt/vertica/bin/vbr.py -t copycluster --config- file /home/dbadmin/backup_script/copy_cluster.ini
Configured backup directories in Mapping are ignored for copycluster.
Starting copy of database Agoda_DW.
Participating nodes: v_agoda_dw_node0001, v_agoda_dw_node0002, v_agoda_dw_node0003.
Snapshotting database.
Snapshot complete.
Syncing data to destination cluster.
[==================================================] 100%
Reinitializing destination catalog.
While bootstrapping catalog: Errors found in SAL or Catalog, check log for details.
While bootstrapping catalog: Errors found in SAL or Catalog, check log for details.
While bootstrapping catalog: Errors found in SAL or Catalog, check log for details.
Error: Catalog bootstrap failed.
Copycluster FAILED.
Hi,
Can you send me the output of the folllowing
From source DB
for host in <list comma seperated ip address>; do ssh $host " find /export/vertica_data1/WH
-name globalsequence.cat -exec cat {} \; " ; done
From target DB
for host in<list comma seperated ip address>; do
ssh $host " find /export/vertica_data1/WH -name globalsequence.cat -exec
cat {} \; " ; done
Sruthi
Hi
Here is the result :
Source:
[dbadmin@bk-vtgpoc-1001 ~]$ for host in 10.120.3.134 10.120.3.135 10.120.3.136; do ssh $host " find /export/vertica_data1/WH -name globalsequence.cat -exec cat {} \; " ; done
find: `/export/vertica_data1/WH': No such file or directory
find: `/export/vertica_data1/WH': No such file or directory
find: `/export/vertica_data1/WH': No such file or directory
Destination
for host in 10.120.1.131 10.120.1.132 10.120.1.133; do ssh $host " find /export/vertica_data1/WH -name globalsequence.cat -exec cat {} \; " ; done
find: `/export/vertica_data1/WH': No such file or directory
CentOS release 6.5 (Final)
Kernel 2.6.32-431.el6.x86_64 on an x86_64
find: `/export/vertica_data1/WH': No such file or directory
CentOS release 6.5 (Final)
Kernel 2.6.32-431.el6.x86_64 on an x86_64
find: `/export/vertica_data1/WH': No such file or directory
Hi,
Is the vertica version same on both clusters?
Thank you,
Sruthi
same version. ^^
Hi,
Looking at the bootstrap-catalog.log, I suspect some file corruption on the target database.
Is this something you can try
1. Drop the target database.
2. Create a new db that meets all the copycluster requirements.
3. Shutdown the target db.
4. Try copycluster from the source.
Thanks
Gayatri
Hi ,
- I try to set up that you recommend . It still fail and show message error about bootstrap fail.
- I try to copy vbr.py file version 7.1 (vbr file from another server not involve testing cluster) and replace it on same location.
then run /opt/vertica/bin/vbr.py -t copycluster --config-file /home/dbadmin/backup_script/copy_cluster.ini
It shows :
Copying...
[==================================================] 100%
All child processes terminated successfully.
copycluster done!
after check the data on destination, data updated
I'm not sure that It is the bug of vbr.py script version 7.2 or not !!
thanks