Horton

Re: [PATCH] random_normal function

Andrei Lepikhov <a.lepikhov@postgrespro.ru>

View thread
From: Andrey Lepikhov <a.lepikhov@postgrespro.ru>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Dean Rasheed <dean.a.rasheed@gmail.com>, Fabien COELHO <coelho@cri.ensmp.fr>, Michael Paquier <michael@paquier.xyz>, Justin Pryzby <pryzby@telsasoft.com>, pgsql-hackers@postgresql.org, Paul Ramsey <pramsey@cleverelephant.ca>
Date: 2023-01-19T06:48:20Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Round off random_normal() test results one more decimal place.

  2. Remove pg_regress' never-documented "ignore" feature.

  3. Upgrade the random.sql regression test.

  4. Invent random_normal() to provide normally-distributed random numbers.

On 1/19/23 11:01, Tom Lane wrote:
> Andrey Lepikhov <a.lepikhov@postgrespro.ru> writes:
>> On 1/9/23 23:52, Tom Lane wrote:
>>> BTW, if this does bring the probability of failure down to the
>>> one-in-a-billion range, I think we could also nuke the whole
>>> "ignore:" business, simplifying pg_regress and allowing the
>>> random test to be run in parallel with others.
> 
>> We have used the pg_sleep() function to interrupt a query at certain
>> execution phase. But on some platforms, especially in containers, the
>> query can vary execution time in so widely that the pg_sleep() timeout,
>> required to get rid of dependency on a query execution time, has become
>> unacceptable. So, the "ignore" option was the best choice.
> 
> But does such a test have any actual value?  If your test infrastructure
> ignores the result, what makes you think you'd notice if the test did
> indeed detect a problem?
Yes, it is good to catch SEGFAULTs and assertions which may be frequent 
because of a logic complexity in the case of timeouts.

> 
> I think "ignore:" was a kluge we put in twenty-plus years ago when our
> testing standards were a lot lower, and it's way past time we got
> rid of it.
Ok, I will try to invent alternative way for deep (and stable) testing 
of timeouts. Thank you for the answer.

-- 
Regards
Andrey Lepikhov
Postgres Professional