# Upgrade from 1.7.4 to 2.3.2 fails, primary shards are not getting restored **URL:** \ **Category:** Elasticsearch\ **Created:** [May 11, 2016, 9:22am UTC](https://discuss.elastic.co/t/upgrade-from-1-7-4-to-2-3-2-fails-primary-shards-are-not-getting-restored/49745 "2016-05-11T09:22:45Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![sweetest.sj](https://avatars.discourse-cdn.com/v4/letter/s/858c86/32.png) [@sweetest.sj](https://discuss.elastic.co/u/sweetest.sj)\ **Post date:** [May 11, 2016, 9:22am UTC](https://discuss.elastic.co/t/upgrade-from-1-7-4-to-2-3-2-fails-primary-shards-are-not-getting-restored/49745/1 "2016-05-11T09:22:45Z")
when I restarted cluster, some of the primaries are properly restored, but some recovery task tasks hang they are in the 'TRANSLOG' stage and are staying there for more than an hour, which is not expected since I run \_flush before restart meaning there should be no remaining translog. hot threads api shows below ``` 0.0% (97.7micros out of 500ms) cpu usage by thread 'elasticsearch[IP][transport_client_timer][T#1]{Hashed wheel timer #1}' 10/10 snapshots sharing following 5 elements java.lang.Thread.sleep(Native Method) org.jboss.netty.util.HashedWheelTimer$Worker.waitForNextTick(HashedWheelTimer.java:445) org.jboss.netty.util.HashedWheelTimer$Worker.run(HashedWheelTimer.java:364) org.jboss.netty.util.ThreadRenamingRunnable.run(ThreadRenamingRunnable.java:108) ``` while /\_cat/recovery shows below INDEX\_NAME 9 1688129 store translog 10.96.250.224 10.96.250.224 n/a n/a 0 100.0% 0 100.0% 111 5087612156 0 -1.0% -1 ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 5, 2017, 10:52pm UTC](https://discuss.elastic.co/t/upgrade-from-1-7-4-to-2-3-2-fails-primary-shards-are-not-getting-restored/49745/2 "2017-07-05T22:52:32Z")
Closed on July 5, 2017, 10:52pm UTC