[PDB Tech] peeringdb sync error

[Stefan Pratter]

 > I just upgraded to 0.4.4.1, same problem with reference to org id 14675.

Upgrading the version most likely wouldn't fix already existing bad 
data, out of curiosity what version were you running before the upgrade?

Without knowing the exact reason why import for the specified org failed 
at some point, this seems like a likely candidate to address the cause: 
https://github.com/peeringdb/peeringdb/issues/32

 > I am not ready to do this, as there's a live system feeding from this
 > data. I suppose I may have to resort to that.

Understood - in case you are looking to fix your db without wiping it, 
the error you are seeing is from the specified org (14675) missing from 
your db.

https://www.peeringdb.com/api/org/14675?depth=1&pretty

Since it only has it's name set, it should be easy to manually insert 
that into your db, and see if that fixes it. Make sure to use the EXACT 
id, name, updated and created values.

If you do that and there are more errors after this, a clean sync would 
most likely be the best way to go.

Stefan


On 02/08/16 19:48, Paul Vlaar wrote:
> On 2/8/16 17:52, Stefan Pratter wrote:
>> Cannot reproduce the issue on my local sync - hadn't run it since Jul
>> 31st and updated without errors just now.
>>
>> I'd appreciate if you could answer these questions to help us identify
>> the cause:
>>
>> 1. which version of py-peeringdb are you running
>
> I just upgraded to 0.4.4.1, same problem with reference to org id 14675.
>
>> 2. "org last update 1470142503 0 changed" indicates that you ran the
>> script earlier today, did that produce any errors? particularly with >
> the org sync? if so which error?
>
> Unsure, since I run this every hour from cron and hadn't look at the
> output for a few weeks (and is over-written on each run).
>
>> 3. Does the problem persist if you do it on a fresh db (eg. after
>> running the drop_tables command, WARNING: this will delete all
>> existing data)
>
> I am not ready to do this, as there's a live system feeding from this
> data. I suppose I may have to resort to that.
>
> Thanks,
>
> 	~paul
> _______________________________________________
> Pdb-tech mailing list
> Pdb-tech at lists.peeringdb.com
> http://lists.peeringdb.com/cgi-bin/mailman/listinfo/pdb-tech
>