Well, until it's confirmed what the exact problem is, or what causes
Tom's issue, solving it is hard for anyone. It could be a time-out
limit.

The two authors showing is because that part is not reindexed yet. New
edits are indexed right away, but updates from before the 'New Index'
haven't been touched.
My favourite example of that is
<http://openlibrary.org/search/authors?q=shirley+conference+manchester+1977>.
All these 'authors' have been merged, but you couldn't tell until you
try to merge them.

Ben

On 13 March 2013 00:43, Karen Coyle <[email protected]> wrote:
> Ben, good thinking. I looked at an author to merge (Terry Pratchett),
> found two author entries, clicked on merge, and then discovered that
> they had already been merged (by me) before. So I don't know why two
> versions are displaying. I also tried the Schulz query and got the same
> error. Problem is, this may take some intense de-bugging, and I don't
> think there's anyone with the knowledge and time at the moment. Once
> Anand finishes his transfer of the backend to the Archive, and indexing
> is happening again, it might be good to look at this. Tom M, would you
> be willing to file a bug report so that we don't forget it?
>
> Thanks,
> kc
>
> On 3/12/13 4:13 PM, Ben Companjen wrote:
>> It's not a direct answer to your question, but are you sure the
>> authors haven't been merged before (causing the error)?
>>
>> Whenever I encountered the fail message, I imagined some author(s) had
>> too many works; the author references couldn't be updated before a
>> timeout. Doing the merge in steps worked well in (most of) these
>> cases.
>>
>> In some cases I think the merge continued despite the error message.
>> Checking back later revealed that.
>>
>> I have wondered whether the whole merge operation is transaction safe
>> (either succes: all changes are committed, or failure: none are
>> committed), but as I haven't encountered errors after many merges, I
>> stopped wondering.
>>
>> Ben
>>
>> On 10 March 2013 00:30, Tom Morris <[email protected]> wrote:
>>> When I get an author merge failure, is it likely that it will succeed at
>>> some point in the future if I retry it or is a permanent error?
>>>
>>> A recent example that I tried was this one for Charles Schultz:
>>>
>>> http://openlibrary.org/authors/OL4361397A/Charles_M._Schulz?merge=true&duplicates=OL6889715A,OL2627137A,OL3120928A,OL2627572A,OL3235100A,OL4766397A,OL5596735A,OL3120936A,OL3297636A,OL3351492A,OL3473907A,OL3474707A,OL3763665A,OL3948598A
>>>
>>> Tom
>>>
>>> _______________________________________________
>>> Ol-tech mailing list
>>> [email protected]
>>> http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech
>>> To unsubscribe from this mailing list, send email to
>>> [email protected]
>>>
>> _______________________________________________
>> Ol-tech mailing list
>> [email protected]
>> http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech
>> To unsubscribe from this mailing list, send email to 
>> [email protected]
>>
>
> --
> Karen Coyle
> [email protected] http://kcoyle.net
> ph: 1-510-540-7596
> m: 1-510-435-8234
> skype: kcoylenet
> _______________________________________________
> Ol-tech mailing list
> [email protected]
> http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech
> To unsubscribe from this mailing list, send email to 
> [email protected]
_______________________________________________
Ol-tech mailing list
[email protected]
http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech
To unsubscribe from this mailing list, send email to 
[email protected]

Reply via email to