[Postgres-xl-bugs] cache lookup failed for node ERROR when using XL

Pavan Deolasee pavan.deolasee at gmail.com
Thu Mar 10 09:12:04 PST 2016


Just so that I understand correctly, does this error occur only after you
send -INT to processes and restart them? What happens if you set the
cluster from scratch, keep autovacuum on, run your tests without
killing/restarting servers?

Thanks,
Pavan

On Thu, Mar 10, 2016 at 10:38 PM, user <susernameb at gmail.com> wrote:

> I did not see any problems with X2 either using “REL_1_2_STABLE”. Just
> with XL.
>
>
>
> Thanks!
>
>
>
> *From:* Koichi Suzuki [mailto:koichi.dbms at gmail.com]
> *Sent:* Thursday, March 10, 2016 11:30 AM
> *To:* user
> *Cc:* Pavan Deolasee; postgres-xc-bugs
>
> *Subject:* Re: [Postgres-xl-bugs] cache lookup failed for node ERROR when
> using XL
>
>
>
> I've not tried to kill process with -INT, but -KILL to mimic sudden crash.
>     After restart, I didn't have any problem (with XC but I expect XL
> behaves the same).
>
>
>
> kill -INT may cause postmaster to do some extra.   Pavan, do you have any
> idea?
>
>
>
> Regards;
>
>
> ---
>
> Koichi Suzuki
>
> https://www.linkedin.com/in/koichidbms
>
>
>
>
>
> 2016-03-10 7:49 GMT-08:00 user <susernameb at gmail.com>:
>
> We use the following commands to kill the cluster
>
>
>
> kill –INT `head data/coord1/postmaster.pid`
>
> kill –INT `head data/dn1/postmaster.pid`
>
> kill –INT `head data/gtm/gtm.pid`
>
>
>
>
>
> *From:* Pavan Deolasee [mailto:pavan.deolasee at gmail.com]
> *Sent:* Wednesday, March 09, 2016 11:25 PM
> *To:* user
> *Cc:* Koichi Suzuki; postgres-xc-bugs
> *Subject:* Re: [Postgres-xl-bugs] cache lookup failed for node ERROR when
> using XL
>
>
>
>
>
> On Wed, Mar 9, 2016 at 4:28 AM, user <susernameb at gmail.com> wrote:
>
> May be this helps in debugging.
>
>
>
> Executing the query “select oid, * from pgxc_node;” on coordinator and on
> data node
>
> Gives different OIDs.
>
>
>
>
>
> Its quite normal to have different OIDs on different nodes. They are
> generated and consumed locally. I wonder if the origin to the problem is
> same as you reported on other thread.
>
>
>
> How do you kill the cluster? Do you use pgxc_ctl to stop the cluster or
> send -9 or some such signal to GTM/postmaster processes?
>
>
>
> Thanks,
>
> Pavan
>
> --
>
>  Pavan Deolasee                   http://www.2ndQuadrant.com/
>  PostgreSQL Development, 24x7 Support, Training & Services
>
>
>



-- 
 Pavan Deolasee                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.postgres-xl.org/private.cgi/postgres-xl-bugs-postgres-xl.org/attachments/20160310/72968f86/attachment.htm>


More information about the postgres-xl-bugs mailing list