All of lore.kernel.org
 help / color / mirror / Atom feed
From: Bob Peterson <rpeterso@redhat.com>
To: cluster-devel.redhat.com
Subject: [Cluster-devel] [DLM PATCH 3/6] DLM: Make consistent error path through tcp_create_listen_sock
Date: Thu, 11 Feb 2016 12:59:10 -0500 (EST)	[thread overview]
Message-ID: <1144926957.21868310.1455213550925.JavaMail.zimbra@redhat.com> (raw)
In-Reply-To: <CAHc6FU7ExQamsAac5iJOz_q2_O4gzT6y1gBY=-qXBziDucJTMg@mail.gmail.com>

----- Original Message -----
> > +out_err:
> > +       sock_release(sock);
> > +       sock = NULL;
> > +       con->sock = NULL;
> 
> Consolidating the error paths makes sense, but con->sock shouldn't be
> set here at all; the caller does that in add_sock().
> 
> Thanks,
> Andreas
> 

Hi Andreas,

I disagree.

The caller doesn't call add_sock() in the error case, which is where
we're setting con->sock to NULL.

Instead, it returns -EADDRINUSE to its caller, dlm_lowcomms_start, which
jumps to label fail_unlisten, which calls close_connection for the con if
it finds it for nodeid 0, then it calls close_connection for it, which
does some things if con->sock is not NULL. That path through the code is
hard to follow, so maybe I'm wrong.

Regards,

Bob Peterson
Red Hat File Systems



  reply	other threads:[~2016-02-11 17:59 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-10 18:55 [Cluster-devel] [DLM PATCH 0/6] Misc DLM Improvements Regarding Socket Errors Bob Peterson
2016-02-10 18:55 ` [Cluster-devel] [DLM PATCH 1/6] DLM: Don't create kernel socket until we have valid node address Bob Peterson
2016-02-10 18:55 ` [Cluster-devel] [DLM PATCH 2/6] DLM: Call original error report when socket is NULL Bob Peterson
2016-02-11 16:43   ` Andreas Gruenbacher
2016-02-10 18:55 ` [Cluster-devel] [DLM PATCH 3/6] DLM: Make consistent error path through tcp_create_listen_sock Bob Peterson
2016-02-11 16:52   ` Andreas Gruenbacher
2016-02-11 17:59     ` Bob Peterson [this message]
2016-02-11 21:09       ` [Cluster-devel] [DLM PATCH 3/6] DLM: Make consistent error path Andreas Gruenbacher
2016-02-10 18:55 ` [Cluster-devel] [DLM PATCH 4/6] DLM: Eliminate useless goto Bob Peterson
2016-02-11 16:53   ` Andreas Gruenbacher
2016-02-10 18:55 ` [Cluster-devel] [DLM PATCH 5/6] DLM: Add locking to protect save callback assignments Bob Peterson
2016-02-11 17:04   ` Andreas Gruenbacher
2016-02-10 18:55 ` [Cluster-devel] [DLM PATCH 6/6] DLM: save / restore all socket callbacks Bob Peterson
2016-02-11 15:31   ` Steven Whitehouse
2016-02-11 16:43     ` [Cluster-devel] [DLM PATCH 6/6][try #2] " Bob Peterson
2016-02-11 17:10       ` Andreas Gruenbacher
2016-02-11 17:05 ` [Cluster-devel] [DLM PATCH 0/6] Misc DLM Improvements Regarding Socket Errors Andreas Gruenbacher
2016-02-11 17:22 ` David Teigland
2016-02-11 18:39   ` Bob Peterson
2016-02-11 18:59     ` David Teigland
2016-02-15 21:16     ` Bob Peterson
2016-02-15 21:24       ` David Teigland

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1144926957.21868310.1455213550925.JavaMail.zimbra@redhat.com \
    --to=rpeterso@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.