pg.ddx.io  pgsql-hackers@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Fujii Masao <masao.fujii@oss.nttdata.com>
Cc: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Cc: pgsql-hackers@lists.postgresql.org
Subject: Re: Review for GetWALAvailability()
Date: Tue, 16 Jun 2020 14:50:47 -0400
Message-ID: <20200616185047.GA19008@alvherre.pgsql> (raw)
In-Reply-To: <ee477f3c-9bff-c9d3-25b0-4e21aafc5e5d@oss.nttdata.com>

On 2020-Jun-17, Fujii Masao wrote:

> While reading InvalidateObsoleteReplicationSlots() code, I found another issue.
> 
> 			ereport(LOG,
> 					(errmsg("terminating walsender %d because replication slot \"%s\" is too far behind",
> 							wspid, NameStr(slotname))));
> 			(void) kill(wspid, SIGTERM);
> 
> wspid indicates the PID of process using the slot. That process can be
> a backend, for example, executing pg_replication_slot_advance().
> So "walsender" in the above log message is not always correct.

Good point.

> 			int			wspid = ReplicationSlotAcquire(NameStr(slotname),
> 													   SAB_Inquire);
> 
> Why do we need to call ReplicationSlotAcquire() here and mark the slot as
> used by the checkpointer? Isn't it enough to check directly the slot's
> active_pid, instead?
> 
> Maybe ReplicationSlotAcquire() is necessary because
> ReplicationSlotRelease() is called later? If so, why do we need to call
> ReplicationSlotRelease()? ISTM that we don't need to do that if the slot's
> active_pid is zero. No?

I think the point here was that in order to modify the slot you have to
acquire it -- it's not valid to modify a slot you don't own.


> +		/*
> +		 * Signal to terminate the process using the replication slot.
> +		 *
> +		 * Try to signal every 100ms until it succeeds.
> +		 */
> +		if (!killed && kill(active_pid, SIGTERM) == 0)
> +			killed = true;
> +		ConditionVariableTimedSleep(&slot->active_cv, 100,
> +									WAIT_EVENT_REPLICATION_SLOT_DROP);
> +	} while (ReplicationSlotIsActive(slot, NULL));

Note that here you're signalling only once and then sleeping many times
in increments of 100ms -- you're not signalling every 100ms as the
comment claims -- unless the signal fails, but you don't really expect
that.  On the contrary, I'd claim that the logic is reversed: if the
signal fails, *then* you should stop signalling.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services





view thread (38+ messages)  latest in thread

Message-ID: <20200616185047.GA19008@alvherre.pgsql>
Permalink:  ../20200616185047.GA19008@alvherre.pgsql/
Also on:    postgresql.org/message-id/20200616185047.GA19008@alvherre.pgsql

 · 

reply

Reply instructions:

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

* Reply to all the recipients from your mail client; the link fills in
  the headers below: reply via email
  Some clients, webmail especially, drop In-Reply-To and References from
  a link, and the reply then starts a new thread: use the mbox below.

  To: pgsql-hackers@postgresql.org
  Cc: alvherre@2ndquadrant.com, masao.fujii@oss.nttdata.com, horikyota.ntt@gmail.com
  Subject: Re: Review for GetWALAvailability()
  In-Reply-To: <20200616185047.GA19008@alvherre.pgsql>
  References: <ee477f3c-9bff-c9d3-25b0-4e21aafc5e5d@oss.nttdata.com> <20200616185047.GA19008@alvherre.pgsql>

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

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox