Re: What Happens to Consistency if I kill a Leader and Startup it again?
Shawn thanks for detailed answer, it explains everything. I think that there is no problem. I will use 4.3. when it is available and if I see a situation something like that I will report. 2013/5/3 Shawn Heisey s...@elyograg.org On 5/2/2013 2:19 PM, Furkan KAMACI wrote: I see that at my admin page: Replication (Slave) Version GenSize Master: 1367307652512 82 778.04 MB Slave: 1367307658862 82 781.05 MB and I started to figure about it so that's why I asked this question. As we've been trying to tell you, the sizes can (and will) be different between replicas on SolrCloud. Also, if you're not running a recent release candidate of 4.3, then the version numbers on the replication screen are misleading. See SOLR-4661 for more details. Your example of version numbers like 100, 90, and 95 wouldn't actually happen, because the version number is based on the current time in milliseconds since 1970-01-01 00:00:00 UTC. If you index after killing the leader, the new leader's version number will be higher than the offline replica. If you can find actual proof of a problem with index updates related to killing the leader, then we can take the bug report and work on fixing it. Here's how you would go about finding proof. It would be easiest to have one shard, but if you want to make sure it's OK with multiple shards, you would have to kill all the leaders. * Start with a functional collection with two replicas. * Index a document with a recognizable ID like A. * Make sure you can find document A. * Kill the leader replica, let's say it was replica1. * Make sure replica2 becomes leader. * Make sure you can find document A. * Index document B. * Start replica1, wait for it to turn green. * Make sure you can still find document B. * Kill the leader again, this time it's replica2. * Make sure you can still find document B. To my knowledge, nobody has reported a real problem with proof. I would imagine that more than one person has done testing like this to make sure that SolrCloud is reliable. Thanks, Shawn
Re: What Happens to Consistency if I kill a Leader and Startup it again?
The leader would not be behind replica because the old leader would not come back and take over the leader role. It would ne just a replica and it would replicate the index from whichever node is the leader. Otis Solr ElasticSearch Support http://sematext.com/ On Apr 29, 2013 5:31 PM, Furkan KAMACI furkankam...@gmail.com wrote: I think about such situation: Let's assume that I am indexing at my SolrCloud. My leader has a version of higher than replica as well (I have one leader and one replica for each shard). If I kill leader, replica will be leader as well. When I startup old leader again it will be a replica for my shard. However I think that leader will have less document than replica and a less version than replica. Does it cause a problem because of leader is behing of replica?
Re: What Happens to Consistency if I kill a Leader and Startup it again?
Thanks for the answer. This is what I try to say: time = t Node A (Leader): version is 100 Node B (Replica): version is 90 time = t+1 Node A (Killing): version is 100 and killed Node B (Replica): version is 90 time = t+2 Node A (Killed): version is 100 and killed Node B (Become Leader): version is 95 (we indexed something) time = t+3 Node A (Started as Replica): version is 100 and live Node B (Leader): version is 95 so I think that leader will behind replica. Is there anything different for such scenario? 2013/5/2 Otis Gospodnetic otis.gospodne...@gmail.com The leader would not be behind replica because the old leader would not come back and take over the leader role. It would ne just a replica and it would replicate the index from whichever node is the leader. Otis Solr ElasticSearch Support http://sematext.com/ On Apr 29, 2013 5:31 PM, Furkan KAMACI furkankam...@gmail.com wrote: I think about such situation: Let's assume that I am indexing at my SolrCloud. My leader has a version of higher than replica as well (I have one leader and one replica for each shard). If I kill leader, replica will be leader as well. When I startup old leader again it will be a replica for my shard. However I think that leader will have less document than replica and a less version than replica. Does it cause a problem because of leader is behing of replica?
Re: What Happens to Consistency if I kill a Leader and Startup it again?
If you're using zookeeper, this should not be allowed to happen (I think). On Thu, May 2, 2013 at 2:12 PM, Furkan KAMACI furkankam...@gmail.comwrote: Thanks for the answer. This is what I try to say: time = t Node A (Leader): version is 100 Node B (Replica): version is 90 time = t+1 Node A (Killing): version is 100 and killed Node B (Replica): version is 90 time = t+2 Node A (Killed): version is 100 and killed Node B (Become Leader): version is 95 (we indexed something) time = t+3 Node A (Started as Replica): version is 100 and live Node B (Leader): version is 95 so I think that leader will behind replica. Is there anything different for such scenario? 2013/5/2 Otis Gospodnetic otis.gospodne...@gmail.com The leader would not be behind replica because the old leader would not come back and take over the leader role. It would ne just a replica and it would replicate the index from whichever node is the leader. Otis Solr ElasticSearch Support http://sematext.com/ On Apr 29, 2013 5:31 PM, Furkan KAMACI furkankam...@gmail.com wrote: I think about such situation: Let's assume that I am indexing at my SolrCloud. My leader has a version of higher than replica as well (I have one leader and one replica for each shard). If I kill leader, replica will be leader as well. When I startup old leader again it will be a replica for my shard. However I think that leader will have less document than replica and a less version than replica. Does it cause a problem because of leader is behing of replica?
Re: What Happens to Consistency if I kill a Leader and Startup it again?
Hi, Can you actually make this happen? Otis Solr ElasticSearch Support http://sematext.com/ On May 2, 2013 8:12 AM, Furkan KAMACI furkankam...@gmail.com wrote: Thanks for the answer. This is what I try to say: time = t Node A (Leader): version is 100 Node B (Replica): version is 90 time = t+1 Node A (Killing): version is 100 and killed Node B (Replica): version is 90 time = t+2 Node A (Killed): version is 100 and killed Node B (Become Leader): version is 95 (we indexed something) time = t+3 Node A (Started as Replica): version is 100 and live Node B (Leader): version is 95 so I think that leader will behind replica. Is there anything different for such scenario? 2013/5/2 Otis Gospodnetic otis.gospodne...@gmail.com The leader would not be behind replica because the old leader would not come back and take over the leader role. It would ne just a replica and it would replicate the index from whichever node is the leader. Otis Solr ElasticSearch Support http://sematext.com/ On Apr 29, 2013 5:31 PM, Furkan KAMACI furkankam...@gmail.com wrote: I think about such situation: Let's assume that I am indexing at my SolrCloud. My leader has a version of higher than replica as well (I have one leader and one replica for each shard). If I kill leader, replica will be leader as well. When I startup old leader again it will be a replica for my shard. However I think that leader will have less document than replica and a less version than replica. Does it cause a problem because of leader is behing of replica?
Re: What Happens to Consistency if I kill a Leader and Startup it again?
Hi Otis; I see that at my admin page: Replication (Slave) Version GenSize Master: 1367307652512 82 778.04 MB Slave: 1367307658862 82 781.05 MB and I started to figure about it so that's why I asked this question. 2013/5/2 Otis Gospodnetic otis.gospodne...@gmail.com Hi, Can you actually make this happen? Otis Solr ElasticSearch Support http://sematext.com/ On May 2, 2013 8:12 AM, Furkan KAMACI furkankam...@gmail.com wrote: Thanks for the answer. This is what I try to say: time = t Node A (Leader): version is 100 Node B (Replica): version is 90 time = t+1 Node A (Killing): version is 100 and killed Node B (Replica): version is 90 time = t+2 Node A (Killed): version is 100 and killed Node B (Become Leader): version is 95 (we indexed something) time = t+3 Node A (Started as Replica): version is 100 and live Node B (Leader): version is 95 so I think that leader will behind replica. Is there anything different for such scenario? 2013/5/2 Otis Gospodnetic otis.gospodne...@gmail.com The leader would not be behind replica because the old leader would not come back and take over the leader role. It would ne just a replica and it would replicate the index from whichever node is the leader. Otis Solr ElasticSearch Support http://sematext.com/ On Apr 29, 2013 5:31 PM, Furkan KAMACI furkankam...@gmail.com wrote: I think about such situation: Let's assume that I am indexing at my SolrCloud. My leader has a version of higher than replica as well (I have one leader and one replica for each shard). If I kill leader, replica will be leader as well. When I startup old leader again it will be a replica for my shard. However I think that leader will have less document than replica and a less version than replica. Does it cause a problem because of leader is behing of replica?
Re: What Happens to Consistency if I kill a Leader and Startup it again?
On 5/2/2013 2:19 PM, Furkan KAMACI wrote: I see that at my admin page: Replication (Slave) Version GenSize Master: 1367307652512 82 778.04 MB Slave: 1367307658862 82 781.05 MB and I started to figure about it so that's why I asked this question. As we've been trying to tell you, the sizes can (and will) be different between replicas on SolrCloud. Also, if you're not running a recent release candidate of 4.3, then the version numbers on the replication screen are misleading. See SOLR-4661 for more details. Your example of version numbers like 100, 90, and 95 wouldn't actually happen, because the version number is based on the current time in milliseconds since 1970-01-01 00:00:00 UTC. If you index after killing the leader, the new leader's version number will be higher than the offline replica. If you can find actual proof of a problem with index updates related to killing the leader, then we can take the bug report and work on fixing it. Here's how you would go about finding proof. It would be easiest to have one shard, but if you want to make sure it's OK with multiple shards, you would have to kill all the leaders. * Start with a functional collection with two replicas. * Index a document with a recognizable ID like A. * Make sure you can find document A. * Kill the leader replica, let's say it was replica1. * Make sure replica2 becomes leader. * Make sure you can find document A. * Index document B. * Start replica1, wait for it to turn green. * Make sure you can still find document B. * Kill the leader again, this time it's replica2. * Make sure you can still find document B. To my knowledge, nobody has reported a real problem with proof. I would imagine that more than one person has done testing like this to make sure that SolrCloud is reliable. Thanks, Shawn
