Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.96) (envelope-from ) id 1vguE8-006YIc-2D for pgsql-hackers@arkaria.postgresql.org; Sat, 17 Jan 2026 00:31:41 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.96) (envelope-from ) id 1vguE7-005omC-1t for pgsql-hackers@arkaria.postgresql.org; Sat, 17 Jan 2026 00:31:39 +0000 Received: from makus.postgresql.org ([2001:4800:3e1:1::229]) by malur.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.96) (envelope-from ) id 1vguE7-005olv-0w for pgsql-hackers@lists.postgresql.org; Sat, 17 Jan 2026 00:31:39 +0000 Received: from goedel.df7cb.de ([49.13.90.212]) by makus.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.96) (envelope-from ) id 1vguE4-000qJB-0A for pgsql-hackers@lists.postgresql.org; Sat, 17 Jan 2026 00:31:38 +0000 Received: from msg.df7cb.de (unknown [IPv6:2a02:908:1472:9340:f0ad:fc6e:9c86:f1dc]) by goedel.df7cb.de (Postfix) with ESMTPSA id 8CF451E1; Sat, 17 Jan 2026 00:31:33 +0000 (UTC) Date: Sat, 17 Jan 2026 01:31:32 +0100 From: Christoph Berg To: Tomas Vondra Cc: Jakub Wartak , pgsql-hackers@lists.postgresql.org Subject: Re: failed NUMA pages inquiry status: Operation not permitted Message-ID: References: <183fe9ab-6010-4cca-b648-1deca332ce2a@vondra.me> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Archived-At: Precedence: bulk Re: Tomas Vondra > Here's WIP fix for the root cause, i.e. handling status -2 in the two > views querying NUMA node for memory pages: Thanks! > I'm not sure if we need to mention this in the docs. It probably should > mention the column can be NULL, which means "unknown node". We could simply say The returned value can be NULL if the NUMA node cannot be determined, e.g. when the page has been swapped out. > Subject: [PATCH] Handle ENOENT status when querying NUMA node ... > We touch the memory before querying the status, but there is no > guarangee it won't be moved to swap in between. We do the touching only ^ guarantee Christoph