| From: | Thomas Munro <thomas(dot)munro(at)enterprisedb(dot)com> |
|---|---|
| To: | Greg k <gregg(dot)kay(at)gmail(dot)com> |
| Cc: | Stephen Frost <sfrost(at)snowman(dot)net>, "pgsql-bugs(at)lists(dot)postgresql(dot)org" <pgsql-bugs(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: LDAP authentication fails with concurrent create extensions |
| Date: | 2018-04-20 02:27:45 |
| Message-ID: | CAEepm=0YiMV-Orprr8btb1b+hpjJfTO=UyNvtBfLrXLARzBvfQ@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
On Fri, Apr 20, 2018 at 2:09 PM, Thomas Munro
<thomas(dot)munro(at)enterprisedb(dot)com> wrote:
> On Fri, Apr 20, 2018 at 1:43 PM, Greg k <gregg(dot)kay(at)gmail(dot)com> wrote:
>> Do I need to raise this as an issue or bug?
>
> It seems like we might need a patch on that adds ldap_set_option(ldap,
> LDAP_OPT_RESTART, LDAP_OPT_ON) somewhere in auth.c to tell it to
> restart on EINTR.
I'm a bit confused about that; how are you supposed to do that early
enough to cover all its syscalls if you don't have the handle until
after you've connected? Perhaps CheckLDAPAuth() (or higher up) should
be blocking signals instead, or at least SIGUSR1 (I guess that's what
you're receiving -- can you tell?)
--
Thomas Munro
http://www.enterprisedb.com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | PG Bug reporting form | 2018-04-20 05:31:33 | BUG #15163: Unable to get the OpenGPG data |
| Previous Message | Thomas Munro | 2018-04-20 02:09:55 | Re: LDAP authentication fails with concurrent create extensions |