Re: [Postgres-xc-developers] GTM Standby Crash

[Nikhil S. <ni...@st...>]

>> IMHO, using TopMemoryContext to mean the top context of each thread is
>> pretty confusing. Bad choice of name for the memory context according
>> to me. Maybe we could have avoided this crash if we had used a
>> different name for the context.
>>
>> This "TopMemoryContext" goes away when that thread goes away. So ain't
>> nothing TOP about it.
>
> Well, let me at least try and defend because that's my baby :-) I
> think I chose name TopTransactionContext because I wanted to give a
> thread in GTM as much the same treatment as a process gets in
> Postgres. So I stick to the same names, but invented TopMost to mean
> the context which is global to the GTM process. My idea was and still
> is that we should avoid using TopMost as much as we can because that
> memory leaks will be hard to plug-in. Remember, we expect GTM to run
> as long as any one component of the cluster is running.. which pretty
> much means forever because while any one component of the cluster can
> go down, but not the entire cluster.
>
> But if its causing confusion, I won't mind adding a code commentary to
> explain the difference. Clearly my fault.
>

Thanks for the explanation Pavan. I come from the Postgres source code
background and when I started looking at this problem I looked at the
memory context and thought everything is fine because it said
"TopMemoryContext". That confused me a little bit. Had it said
"ThreadTopContext" it would have been much more readable IMHO.

Regards,
Nikhils
-- 
StormDB - http://www.stormdb.com
The Database Cloud
Postgres-XC Support and Service