Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1kfMJ9-0006JW-O6 for pgsql-performance@arkaria.postgresql.org; Wed, 18 Nov 2020 12:11:31 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.92) (envelope-from ) id 1kfMJ7-0003yy-B1 for pgsql-performance@arkaria.postgresql.org; Wed, 18 Nov 2020 12:11:29 +0000 Received: from makus.postgresql.org ([2001:4800:3e1:1::229]) by malur.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1kfMJ7-0003yl-3z for pgsql-performance@lists.postgresql.org; Wed, 18 Nov 2020 12:11:29 +0000 Received: from mail-wr1-x434.google.com ([2a00:1450:4864:20::434]) by makus.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.92) (envelope-from ) id 1kfMJ4-0000Am-8l for pgsql-performance@postgresql.org; Wed, 18 Nov 2020 12:11:27 +0000 Received: by mail-wr1-x434.google.com with SMTP id p1so1905866wrf.12 for ; Wed, 18 Nov 2020 04:11:25 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cybertec-at.20150623.gappssmtp.com; s=20150623; h=message-id:subject:from:to:cc:date:in-reply-to:references :user-agent:mime-version:content-transfer-encoding; bh=7qDebDKurq4o0qzyzxvr1FY/dJQy91iVZKg24H8oXjw=; b=J76yF+ZvXWIyxtIAV+ruos8hI2ef8UWCxQEPsWl1FhWyNCV3syHdSWit0iy5ikXQkC G7ntUusgitLT9Fk2BWp+bnZiMlvp0veopJhmuFr8r9YeKviTOKuZshexUPSwXrkxE/ry MJB2yBDP3GTku1ozzFH/WHfvzRz8w5+uK1EquBCvRDCafntT+LT3KiN9PeBcJqOoLywL ag9gMrkyNhl+QAmDWzwVysK+jOvts9KxkAqP59IUtZ816u0mEzgAchxR3/RcC4k96nOh xUHuKPXYSGoiBFahpHNpgNuk3lvTW/ICOKUNKdM68W9b/rvery1SRrMGP9Ztt4k4N77B 83VQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:message-id:subject:from:to:cc:date:in-reply-to :references:user-agent:mime-version:content-transfer-encoding; bh=7qDebDKurq4o0qzyzxvr1FY/dJQy91iVZKg24H8oXjw=; b=CHrqpVHTDPMghpitUNj8kGH221NCGkcCaHs2Mny0xxa/VluHi0Bf/Cx7M7t4d1SERf 3GxcFqM86Hvf2w3r8NhP1KBwuV1VhEDqaIhr9Noa9ss8dP3Q9s6dNxHwOAqnTjffwOHz qMg7CxJyzTiq0rBgZw96fdLqPQc3ww74DtYaEC7gX+M6deCo9CQuzHK1o9KjeHkyg9aU tU8uh6AQiL0f85COlXHekQPzaD2DF5aUApJJ0adBqrnK4wLvXifgoKMZ4D/ZTjDVvUfh S2OobKl5lFFN9PuHSqHgydHa0koCrlDFQB/FIHLyxB7NOelitxrgdBy7QtXbHjpelZ/3 +x2Q== X-Gm-Message-State: AOAM532qRwHKAbrvtDmRVuIFZkSFqARpTAA01VjZPggMqf5Tff9aF1ob YaaSt6WObdxzvdu7CRQVELeCPA== X-Google-Smtp-Source: ABdhPJyPfn/zV7bPVAaZE/UjGSTW/sKKXWLlr3Abfl3yLWjKJ9/Nk4irc5bQut90H9yaCWiANlxggg== X-Received: by 2002:adf:f9c5:: with SMTP id w5mr4338665wrr.69.1605701484542; Wed, 18 Nov 2020 04:11:24 -0800 (PST) Received: from localhost.localdomain (217-149-169-56.nat.highway.telekom.at. [217.149.169.56]) by smtp.gmail.com with ESMTPSA id s4sm32236240wro.10.2020.11.18.04.11.24 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 18 Nov 2020 04:11:24 -0800 (PST) Message-ID: Subject: Re: How to prioritise walsender reading from pg_wal over WAL writes? From: Laurenz Albe To: Alexey Bashtanov Cc: pgsql-performance@postgresql.org Date: Wed, 18 Nov 2020 13:11:23 +0100 In-Reply-To: <774ec1b8-c7c6-cfa2-098d-5c6a32dcfc1a@imap.cc> References: <774ec1b8-c7c6-cfa2-098d-5c6a32dcfc1a@imap.cc> Content-Type: text/plain; charset="UTF-8" User-Agent: Evolution 3.36.5 (3.36.5-1.fc32) MIME-Version: 1.0 Content-Transfer-Encoding: 8bit List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Precedence: bulk On Wed, 2020-11-18 at 11:54 +0000, Alexey Bashtanov wrote: > If it helps, here's the details of the hardware config. > The controller is AVAGO MegaRAID SAS 9361-4i, > the SSDs are INTEL SSDSC2KG960G8 (configured as a raid1). > Current scheduler used is deadline. > Currently XFS is mounted without nobarriers, but I'm going to set that when there's high load next time to see how it affects throughput. > > Properties of the array > VD6 Properties : > ============== > Strip Size = 256 KB > Number of Blocks = 1874329600 > VD has Emulated PD = Yes > Span Depth = 1 > Number of Drives Per Span = 2 > Write Cache(initial setting) = WriteBack > Disk Cache Policy = Disk's Default > Encryption = None > Data Protection = Disabled > Active Operations = None > Exposed to OS = Yes > Creation Date = 28-10-2020 > Creation Time = 05:17:16 PM > Emulation type = default > Cachebypass size = Cachebypass-64k > Cachebypass Mode = Cachebypass Intelligent > Is LD Ready for OS Requests = Yes > SCSI NAA Id = 600605b00af03650272c641ca30c3196 Why?? WAL buffers has the most recent information, so that would result in unnecessary delay and I/O. You'd have to hack the code, but I wonder what leads you to this interesting requirement. Yours, Laurenz Albe -- +43-670-6056265 CYBERTEC PostgreSQL International GmbH Gröhrmühlgasse 26, A-2700 Wiener Neustadt Web: https://www.cybertec-postgresql.com