mdtruncate leaking fd.c handles?

Поиск
Список
Период
Сортировка
От Andres Freund
Тема mdtruncate leaking fd.c handles?
Дата
Msg-id 20160908220748.oqh37ukwqqncbl3n@alap3.anarazel.de
обсуждение исходный текст
Ответы Re: mdtruncate leaking fd.c handles?  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
Hi,

Am I missing something or is md.c:mdtruncate() leaking open files?
The relevant piece of code is:
    if (priorblocks > nblocks)    {        /*         * This segment is no longer active (and has already been unlinked
       * from the mdfd_chain). We truncate the file, but do not delete         * it, for reasons explained in the
headercomments.         */        if (FileTruncate(v->mdfd_vfd, 0) < 0)            ereport(ERROR,
(errcode_for_file_access(),                    errmsg("could not truncate file \"%s\": %m",
FilePathName(v->mdfd_vfd))));
        if (!SmgrIsTemp(reln))            register_dirty_segment(reln, forknum, v);        v = v->mdfd_chain;
Assert(ov!= reln->md_fd[forknum]); /* we never drop the 1st                                             * segment */
   pfree(ov);    }
 

note that we're freeing the chain entry, which contains the mdfd_vfd,
without previously calling FileClose().

This only really matters for VACUUM truncate and ON COMMIT TRUNCATE temp
table truncation afaics.  But it seems like we should fix it everywhere
nonetheless.

Greetings,

Andres Freund



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Lucas
Дата:
Сообщение: Re: Preventing deadlock on parallel backup
Следующее
От: Tom Lane
Дата:
Сообщение: Re: mdtruncate leaking fd.c handles?