On Monday, June 27, 2016 at 8:29:23 AM UTC+2, Frank Celler wrote:
>
> We have to debug, why the agents are crashing. I've alerted my colleague 
> who is responsible for the agency. Do have have any core dumps by chance?
>

Maybe crashing isn't the correct wording here, they simply exit without any 
error messages.

Here is a sample output of an agent that started correctly, but then exits 
when another agent is started:

bart@laptop ~ $ arangod --server.endpoint tcp://127.0.0.1:5001 
--server.authentication false --agency.id 0 --agency.size 3 agency1
2016-06-26T15:33:38Z [30589] INFO using SSL options: 
SSL_OP_CIPHER_SERVER_PREFERENCE, SSL_OP_TLS_ROLLBACK_BUG
2016-06-26T15:33:38Z [30589] INFO file-descriptors (nofiles) hard limit is 
4096, soft limit is 1024
2016-06-26T15:33:38Z [30589] INFO ArangoDB 3.0.0 [linux] 64bit, using VPack 
0.1.30, ICU 54.1, V8 5.0.71.39, OpenSSL 1.0.2g-fips  1 Mar 2016
2016-06-26T15:33:38Z [30589] INFO loaded database '_system' from 
'agency1/databases/database-1'
2016-06-26T15:33:38Z [30589] INFO the server has 8 (hyper) cores, using 2 
scheduler thread(s), 8 dispatcher thread(s)
2016-06-26T15:33:38Z [30589] INFO running WAL recovery (1 logfiles)
2016-06-26T15:33:38Z [30589] INFO replaying WAL logfile 
'agency1/journals/logfile-2.db' (1 of 1)
2016-06-26T15:33:38Z [30589] INFO WAL recovery finished successfully
2016-06-26T15:33:38Z [30589] INFO JavaScript using startup 
'/usr/share/arangodb3/js', application '/var/lib/arangodb3-apps'
2016-06-26T15:33:43Z [30589] INFO In database '_system': Found 14 defined 
task(s), 1 task(s) to run
2016-06-26T15:33:43Z [30589] INFO In database '_system': state standalone/
existing, tasks updateUserModels
2016-06-26T15:33:43Z [30589] INFO In database '_system': existing cleanup 
successfully finished
2016-06-26T15:33:44Z [30589] INFO using endpoint 'http+tcp://127.0.0.1:5001' 
for non-encrypted requests
2016-06-26T15:33:44Z [30589] INFO Authentication is turned off
2016-06-26T15:33:44Z [30589] INFO ArangoDB (version 3.0.0 [linux]) is ready 
for business. Have fun!
2016-06-26T15:33:45Z [30589] WARNING {agency} I have a higher term than RPC 
caller.
2016-06-26T15:33:49Z [30589] WARNING {agency} I have a higher term than RPC 
caller.
bart@laptop ~ $

As you can see, it simply exits after printing 2 warnings regarding "higher 
term than RPC caller"

Bart



> Am Sonntag, 26. Juni 2016 18:36:46 UTC+2 schrieb Bart DS:
>>
>> First of all congratulations on the new 3.0 release!
>> Much improved clustering with auto-failover, Velocypack, persistent 
>> indexes, etc are all great new features.
>>
>> Unfortunately I'm having several issues getting the new clustering 
>> working.
>> As discussed in an earlier post, I'm interested in setting up an ArangoDB 
>> cluster with docker images, not relying on Mesos at all (
>> https://groups.google.com/forum/#!topic/arangodb/WHedmT2mkoE)
>> Max promised to publish a blog post on this shortly after the release, so 
>> I hope it is coming soon ;-)
>>
>> The documentation already has a (small) chapter on setting up an ArangoDB 
>> 3 cluster, so I decided to try it out. (
>> https://docs.arangodb.com/3.0/Manual/Deployment/Local.html)
>> However, I'm having several issues getting it working.
>>
>> When I start multiple agents, they are randomly crashing (they exit 
>> without any errors)
>> After restarting the crashed agent(s) until they all stay up (don't crash 
>> anymore), I'm starting the first DBServer
>>
>> However, it hangs on the following line:
>> arangod --server.authentication=false --server.endpoint tcp://
>> 0.0.0.0:8529 --cluster.my-address tcp://127.0.0.1:8529 
>> --cluster.my-local-info db1 --cluster.my-role PRIMARY 
>> --cluster.agency-endpoint tcp://127.0.0.1:5001 primary1
>> 2016-06-26T15:37:06Z [31640] INFO using SSL options: 
>> SSL_OP_CIPHER_SERVER_PREFERENCE, SSL_OP_TLS_ROLLBACK_BUG
>>
>> Any ideas what could be wrong here?
>> Why are the agents crashing and why is the primary server hanging during 
>> startup?
>>
>> I tried on several Linux based systems, running Linux Mint 17.3 x64 with 
>> 8 GB RAM
>>
>> Bart
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"ArangoDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to