ElasticDump process stuck while copying data from 2.4.6 to 6.2.2


We are upgrading our ES servers from 2.4.6 to 6.2.2.
To do so instead of upgrading existing servers, we have taken new machines and installed ES
version 6.2.2.
Now we are copying our existing indices from 2.4.6 machines to 6.2.2 machine using elasticdump.


nohup /root/node_modules/elasticdump/bin/elasticdump --input=http://elasticsearch-alfstg.xxx.yyy.com:9200/alf/ALFData --output=http://isearch-elasticsearch-stg.xxx.yyy.com/software/softwareData --limit=10000 --type=data --maxSockets=5 --scrollTime=15m --debug

This command works fine but intermittently stuck after dumping some docs from source to target.
We have indexed the data till 2,30,000 and after that the process stuck, it might be due to some corrupted data b/w the offset of 2,30,000 to 2,40,000.
So We then moved to the next offset block i.e. from 2,40,000 and continue the elasticdump

Command to start from an offset :

nohup /root/node_modules/elasticdump/bin/elasticdump --input=http://elasticsearch-alfstg.corp.adobe.com:9200/alf/ALFData --output=http://isearch-elasticsearch-stg.corp.adobe.com/software/softwareData --limit=10000 --type=data --maxSockets=5 --scrollTime=15m –offset=240000 --debug &

it again stuck after some time; we then skipped 10k block and restarted the process again.

Any body has any idea why elasticdump stops intermittently? OR any better approach of indices copy.

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.