Nikolay,

Could you look into the Spark related question.

Alex, Ivan,

Any ideas on how we can end up having consistent IDs conflict on restarts?

--
Denis

On Fri, Aug 31, 2018 at 5:54 AM eugene miretsky <[email protected]>
wrote:

> Thanks Denis,
>
> The version is 2.5 and we are indeed using persistence. Full config
> attached.
>
> Another weird thing that happened is that after restarting the node a few
> time it starts properly and joins the cluster. However, when I try to
> create a SQL table (from spark) using template "ga_template" (defined in
> the config) I get the error *Cache doesn't exist. *After some time, with
> the same code and settings, creating the table starts working.
>
> I know it sounds odd, but I have observed both a few times.
>
> Based on the link you sent, the node Id should be automatically picked up
> from the name of the file in persistence directory. Is it possible that the
> pst lock was not being released properly, so when the node was restarted it
> tried to create a new UUID?
>
> Cheers,
> Eugene
>
> On Fri, Aug 31, 2018 at 1:57 AM Denis Magda <[email protected]> wrote:
>
>> Strange, I've never seen consistent IDs collisions before. Are you using
>> Ignite persistence and what's your version? If you scroll to the end of
>> this paragraph, you'll find an explanation on how the IDs are generated:
>>
>> https://apacheignite.readme.io/docs/distributed-persistent-store#section-usage
>>
>> --
>> Denis
>>
>> On Thu, Aug 30, 2018 at 9:10 PM eugene miretsky <
>> [email protected]> wrote:
>>
>>> Hello,
>>>
>>> Is it possible to set a nodeId when restarting a node? How is the id
>>> generated?
>>>
>>> Sometimes after the cluster crashes, when I restart a node I get the
>>> following error: Caused by: class
>>> org.apache.ignite.spi.IgniteSpiException: Failed to add node to topology
>>> because it has the same hash code for partitioned affinity as one of
>>> existing nodes [cacheName=SQL_PUBLIC_GAL3EC2,
>>> existingNodeId=598e3ead-99b8-4c49-b7df-04d578dcbf5f]
>>>
>>> It looks like the node is trying to start with another nodeId, and
>>> cannot because it's old nodeId owns the same partitions.
>>>
>>> Cheers,
>>> Eugene
>>>
>>

Reply via email to