From varnish-bugs at projects.linpro.no Mon Aug 7 09:04:34 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:04:34 -0000 Subject: [Varnish] #23: varnishlog improvement In-Reply-To: <077.ce58d7d130db2a9337cf691c5b31a1b5@projects.linpro.no> References: <077.ce58d7d130db2a9337cf691c5b31a1b5@projects.linpro.no> Message-ID: <086.ddb39f9ba9539373b67d0535793105a6@projects.linpro.no> #23: varnishlog improvement -------------------------+-------------------------------------------------- Reporter: phk | Owner: phk Type: enhancement | Status: new Priority: normal | Milestone: Component: varnishlog | Version: trunk Severity: normal | Resolution: Keywords: | -------------------------+-------------------------------------------------- Changes (by des): * version: => trunk -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:32:04 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:32:04 -0000 Subject: [Varnish] #32: CLI read/writeres In-Reply-To: <077.89a5dec28081e66dce88b1a41ecf5b1e@projects.linpro.no> References: <077.89a5dec28081e66dce88b1a41ecf5b1e@projects.linpro.no> Message-ID: <086.f73aeeae75dcd59c5a502599852d0262@projects.linpro.no> #32: CLI read/writeres ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Varnish 1.0 Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Changes (by des): * version: => trunk * milestone: => Varnish 1.0 -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:32:47 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:32:47 -0000 Subject: [Varnish] #31: slow client timeout In-Reply-To: <077.0477aaf3a3c8bc6e16fde70f028d88ae@projects.linpro.no> References: <077.0477aaf3a3c8bc6e16fde70f028d88ae@projects.linpro.no> Message-ID: <086.1458797e8a24d8297f5ee8ec348ae471@projects.linpro.no> #31: slow client timeout ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Changes (by des): * version: => trunk -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:33:13 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:33:13 -0000 Subject: [Varnish] #28: Doc input: CLI protocol In-Reply-To: <077.64f388c780ac09cff9a67a112dfb8e8d@projects.linpro.no> References: <077.64f388c780ac09cff9a67a112dfb8e8d@projects.linpro.no> Message-ID: <086.fd91d7c496d47fb3fdce444d7dd5992e@projects.linpro.no> #28: Doc input: CLI protocol ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: task | Status: new Priority: normal | Milestone: Component: documentation | Version: trunk Severity: normal | Resolution: Keywords: | ---------------------------+------------------------------------------------ Changes (by des): * version: => trunk -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:33:25 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:33:25 -0000 Subject: [Varnish] #28: Doc input: CLI protocol In-Reply-To: <077.64f388c780ac09cff9a67a112dfb8e8d@projects.linpro.no> References: <077.64f388c780ac09cff9a67a112dfb8e8d@projects.linpro.no> Message-ID: <086.0ecacb55ecbe86ecafba71ad014d8fc3@projects.linpro.no> #28: Doc input: CLI protocol ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: task | Status: new Priority: normal | Milestone: Varnish 1.0 Component: documentation | Version: trunk Severity: normal | Resolution: Keywords: | ---------------------------+------------------------------------------------ Changes (by des): * milestone: => Varnish 1.0 -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:33:45 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:33:45 -0000 Subject: [Varnish] #27: make workspace tweakable In-Reply-To: <077.f0818d299d12c5c8a476fd2a50561e96@projects.linpro.no> References: <077.f0818d299d12c5c8a476fd2a50561e96@projects.linpro.no> Message-ID: <086.a11f0b4813b3168100632669ca3be7fd@projects.linpro.no> #27: make workspace tweakable -------------------------+-------------------------------------------------- Reporter: phk | Owner: phk Type: enhancement | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | -------------------------+-------------------------------------------------- Changes (by des): * version: => trunk -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:34:19 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:34:19 -0000 Subject: [Varnish] #21: Thoughts about the prefetcher In-Reply-To: <077.7be163059f9973c967a094f90c6a71cd@projects.linpro.no> References: <077.7be163059f9973c967a094f90c6a71cd@projects.linpro.no> Message-ID: <086.af85ede0f12c8478ec9d1a039598bd15@projects.linpro.no> #21: Thoughts about the prefetcher ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: enhancement | Status: new Priority: normal | Milestone: Component: documentation | Version: trunk Severity: normal | Resolution: Keywords: | ---------------------------+------------------------------------------------ Changes (by des): * version: => trunk -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:34:35 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:34:35 -0000 Subject: [Varnish] #29: VCL unload/discard leaks In-Reply-To: <077.97ab360db6e4811321b85a639efa60ed@projects.linpro.no> References: <077.97ab360db6e4811321b85a639efa60ed@projects.linpro.no> Message-ID: <086.657ea81af5662cf59c60db26bf1b0eff@projects.linpro.no> #29: VCL unload/discard leaks ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: low | Milestone: Varnish 1.0 Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Changes (by des): * version: => trunk * milestone: => Varnish 1.0 -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:35:02 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:35:02 -0000 Subject: [Varnish] #26: Various minor issues for consideration In-Reply-To: <077.a78b5cc5796b4bd44af177d5d9ebddc1@projects.linpro.no> References: <077.a78b5cc5796b4bd44af177d5d9ebddc1@projects.linpro.no> Message-ID: <086.6f0d67b6aceaca715b5e138cde1d75b0@projects.linpro.no> #26: Various minor issues for consideration -------------------------+-------------------------------------------------- Reporter: phk | Owner: phk Type: enhancement | Status: new Priority: low | Milestone: Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | -------------------------+-------------------------------------------------- Changes (by des): * version: => trunk -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:35:25 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:35:25 -0000 Subject: [Varnish] #24: VCL compiler memory leaks In-Reply-To: <077.a888fdc3cab823abcc30a5a0e6b4dfd5@projects.linpro.no> References: <077.a888fdc3cab823abcc30a5a0e6b4dfd5@projects.linpro.no> Message-ID: <086.d1b7d67ddbdb7189186389e8d3888d6f@projects.linpro.no> #24: VCL compiler memory leaks ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: lowest | Milestone: Component: varnishd | Version: trunk Severity: trivial | Resolution: Keywords: | ----------------------+----------------------------------------------------- Changes (by des): * version: => trunk -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 09:36:04 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 09:36:04 -0000 Subject: [Varnish] #21: Thoughts about the prefetcher In-Reply-To: <077.7be163059f9973c967a094f90c6a71cd@projects.linpro.no> References: <077.7be163059f9973c967a094f90c6a71cd@projects.linpro.no> Message-ID: <086.2b64a34d6c55b4702137b7f2fac6cf3d@projects.linpro.no> #21: Thoughts about the prefetcher ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: enhancement | Status: new Priority: normal | Milestone: Varnish 1.0 Component: documentation | Version: trunk Severity: normal | Resolution: Keywords: | ---------------------------+------------------------------------------------ Changes (by des): * milestone: => Varnish 1.0 -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 17:39:33 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 17:39:33 -0000 Subject: [Varnish] #32: CLI read/writeres In-Reply-To: <077.89a5dec28081e66dce88b1a41ecf5b1e@projects.linpro.no> References: <077.89a5dec28081e66dce88b1a41ecf5b1e@projects.linpro.no> Message-ID: <086.86b1a9a7befe60b79eaa9e9a8a56e18c@projects.linpro.no> #32: CLI read/writeres ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: closed Priority: normal | Milestone: Varnish 1.0 Component: varnishd | Version: trunk Severity: normal | Resolution: fixed Keywords: | ----------------------+----------------------------------------------------- Changes (by phk): * status: new => closed * resolution: => fixed -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Mon Aug 7 17:41:12 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 07 Aug 2006 17:41:12 -0000 Subject: [Varnish] #31: slow client timeout In-Reply-To: <077.0477aaf3a3c8bc6e16fde70f028d88ae@projects.linpro.no> References: <077.0477aaf3a3c8bc6e16fde70f028d88ae@projects.linpro.no> Message-ID: <086.e6f32652ddf2b7be5249176e86abf78a@projects.linpro.no> #31: slow client timeout ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Comment (by phk): Using SO_SNDTIMEO seems to fix this. Sendfile(2) not used at present, need to add it back. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Tue Aug 8 08:01:14 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Tue, 08 Aug 2006 08:01:14 -0000 Subject: [Varnish] #33: Marketing Fodder: 18 syscalls per request Message-ID: <077.3e6f181d2a5bf2b024a3cedcad489655@projects.linpro.no> #33: Marketing Fodder: 18 syscalls per request ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: defect | Status: new Priority: normal | Milestone: Component: documentation | Version: Severity: normal | Keywords: ---------------------------+------------------------------------------------ As far as I can tell, a hit costs only 18 system calls on SC21. Misses take more. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Tue Aug 8 09:06:02 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Tue, 08 Aug 2006 09:06:02 -0000 Subject: [Varnish] #33: Marketing Fodder: 18 syscalls per request In-Reply-To: <077.3e6f181d2a5bf2b024a3cedcad489655@projects.linpro.no> References: <077.3e6f181d2a5bf2b024a3cedcad489655@projects.linpro.no> Message-ID: <086.c55f860eeee8636de379aece3fa00351@projects.linpro.no> #33: Marketing Fodder: 18 syscalls per request ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: defect | Status: new Priority: normal | Milestone: Component: documentation | Version: Severity: normal | Resolution: Keywords: | ---------------------------+------------------------------------------------ Comment (by phk): And come to think about it, it costs us two syscalls to find out how many we do and at least one or two to report it so the real number is lower. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Wed Aug 9 11:41:50 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Wed, 09 Aug 2006 11:41:50 -0000 Subject: [Varnish] #34: req.http.host -> req.host Message-ID: <077.c207b47aaa1e01a1472e608bccb9e170@projects.linpro.no> #34: req.http.host -> req.host -------------------------+-------------------------------------------------- Reporter: des | Owner: des Type: enhancement | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Keywords: -------------------------+-------------------------------------------------- [from http://projects.linpro.no/pipermail/varnish-dev/2006- August/000368.html] What I want to do in the long run is: * define HTTP_HDR_HOST and the corresponding req.host variable to be used instead of req.http.host. * add code to cache_http.c to dissect and normalize the request URL, and set the initial value of req.host based on the host part of the request URL and / or the Host: header * make req.host and req.url writable so vcl_recv can perform further normalization if needed (e.g. rewrite vg.no to www.vg.no) -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Fri Aug 11 14:20:28 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Fri, 11 Aug 2006 14:20:28 -0000 Subject: [Varnish] #34: req.http.host -> req.host In-Reply-To: <077.c207b47aaa1e01a1472e608bccb9e170@projects.linpro.no> References: <077.c207b47aaa1e01a1472e608bccb9e170@projects.linpro.no> Message-ID: <086.e3cdda1957b2586bf5fe19c79cb5aca8@projects.linpro.no> #34: req.http.host -> req.host -------------------------+-------------------------------------------------- Reporter: des | Owner: des Type: enhancement | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | -------------------------+-------------------------------------------------- Comment (by phk): As I said in email already, I disagree with this direction. Host is a HTTP header, it is not a HTTP field, and it complicates the code for us if we try to pretend otherwise. Furthermore, I have yet to see a single request with the hostname in the URL field. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Fri Aug 11 14:42:25 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Fri, 11 Aug 2006 14:42:25 -0000 Subject: [Varnish] #11: Random memory corruption In-Reply-To: <077.c881c36e47621b4de824d8956c6f981e@projects.linpro.no> References: <077.c881c36e47621b4de824d8956c6f981e@projects.linpro.no> Message-ID: <086.e7046edb033bf9c21abefb62b76e2ef8@projects.linpro.no> #11: Random memory corruption ----------------------+----------------------------------------------------- Reporter: des | Owner: phk Type: defect | Status: closed Priority: normal | Milestone: Varnish 1.0 Component: varnishd | Version: trunk Severity: major | Resolution: fixed Keywords: | ----------------------+----------------------------------------------------- Changes (by phk): * status: assigned => closed * resolution: => fixed Comment: I hereby declare this issue resolved and the ticket closed. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Fri Aug 11 19:33:17 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Fri, 11 Aug 2006 19:33:17 -0000 Subject: [Varnish] #4: VCL error implementation In-Reply-To: <077.9878814473f445bc77c10fbaf1742bf9@projects.linpro.no> References: <077.9878814473f445bc77c10fbaf1742bf9@projects.linpro.no> Message-ID: <086.f1774e151a8677120ac65d3b6afb2294@projects.linpro.no> #4: VCL error implementation ----------------------+----------------------------------------------------- Reporter: des | Owner: des Type: defect | Status: new Priority: normal | Milestone: Varnish 1.0 Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Changes (by des): * owner: phk => des Comment: comment from phk on #varnish: 20:54 < phk> min tanke var at lave et indbygget dokument for hver fejl (vsb_printf(...)) og saa lave det saa tar-storage kan levere alternative dokumenter -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sun Aug 13 12:03:12 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sun, 13 Aug 2006 12:03:12 -0000 Subject: [Varnish] #4: VCL error implementation In-Reply-To: <077.9878814473f445bc77c10fbaf1742bf9@projects.linpro.no> References: <077.9878814473f445bc77c10fbaf1742bf9@projects.linpro.no> Message-ID: <086.bfb41a82f6da8fbb80bbaa3d51130fbd@projects.linpro.no> #4: VCL error implementation ----------------------+----------------------------------------------------- Reporter: des | Owner: des Type: defect | Status: assigned Priority: normal | Milestone: Varnish 1.0 Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Changes (by des): * status: new => assigned -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Wed Aug 16 11:38:09 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Wed, 16 Aug 2006 11:38:09 -0000 Subject: [Varnish] #4: VCL error implementation In-Reply-To: <077.9878814473f445bc77c10fbaf1742bf9@projects.linpro.no> References: <077.9878814473f445bc77c10fbaf1742bf9@projects.linpro.no> Message-ID: <086.699f7acd4cc8842d819b4561784182b2@projects.linpro.no> #4: VCL error implementation ----------------------+----------------------------------------------------- Reporter: des | Owner: des Type: defect | Status: closed Priority: normal | Milestone: Varnish 1.0 Component: varnishd | Version: trunk Severity: normal | Resolution: fixed Keywords: | ----------------------+----------------------------------------------------- Changes (by des): * status: assigned => closed * resolution: => fixed Comment: Basic error functionality is now in place. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Wed Aug 16 11:39:13 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Wed, 16 Aug 2006 11:39:13 -0000 Subject: [Varnish] #35: Support for external error documents Message-ID: <077.ac66fa13c12e2e5001458b2373248b3b@projects.linpro.no> #35: Support for external error documents -------------------------+-------------------------------------------------- Reporter: des | Owner: phk Type: enhancement | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Keywords: -------------------------+-------------------------------------------------- Extend the VCL error function to support custom error documents. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Wed Aug 16 11:40:29 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Wed, 16 Aug 2006 11:40:29 -0000 Subject: [Varnish] #21: Thoughts about the prefetcher In-Reply-To: <077.7be163059f9973c967a094f90c6a71cd@projects.linpro.no> References: <077.7be163059f9973c967a094f90c6a71cd@projects.linpro.no> Message-ID: <086.3927a9b6246c50993958664d88e82420@projects.linpro.no> #21: Thoughts about the prefetcher ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: enhancement | Status: new Priority: normal | Milestone: Component: documentation | Version: trunk Severity: normal | Resolution: Keywords: | ---------------------------+------------------------------------------------ Changes (by des): * milestone: Varnish 1.0 => Comment: Prefetching is not currently considered a requirement for 1.0. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Wed Aug 16 11:41:03 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Wed, 16 Aug 2006 11:41:03 -0000 Subject: [Varnish] #21: Thoughts about the prefetcher In-Reply-To: <077.7be163059f9973c967a094f90c6a71cd@projects.linpro.no> References: <077.7be163059f9973c967a094f90c6a71cd@projects.linpro.no> Message-ID: <086.f8ec42475e4ee37a574f53d5a79d4ae1@projects.linpro.no> #21: Thoughts about the prefetcher -------------------------+-------------------------------------------------- Reporter: phk | Owner: phk Type: enhancement | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | -------------------------+-------------------------------------------------- Changes (by des): * owner: des => phk * component: documentation => varnishd -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Wed Aug 16 11:47:08 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Wed, 16 Aug 2006 11:47:08 -0000 Subject: [Varnish] #28: Doc input: CLI protocol In-Reply-To: <077.64f388c780ac09cff9a67a112dfb8e8d@projects.linpro.no> References: <077.64f388c780ac09cff9a67a112dfb8e8d@projects.linpro.no> Message-ID: <086.74778382da466799875dc2d8d5fde2fe@projects.linpro.no> #28: Doc input: CLI protocol ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: documentation | Status: new Priority: normal | Milestone: Varnish 1.0 Component: documentation | Version: trunk Severity: normal | Resolution: Keywords: | ---------------------------+------------------------------------------------ Changes (by des): * type: task => documentation -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Wed Aug 16 11:48:10 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Wed, 16 Aug 2006 11:48:10 -0000 Subject: [Varnish] #33: Marketing Fodder: 18 syscalls per request In-Reply-To: <077.3e6f181d2a5bf2b024a3cedcad489655@projects.linpro.no> References: <077.3e6f181d2a5bf2b024a3cedcad489655@projects.linpro.no> Message-ID: <086.ca074f5b3c062bd4c7e208594cceaab3@projects.linpro.no> #33: Marketing Fodder: 18 syscalls per request ---------------------------+------------------------------------------------ Reporter: phk | Owner: des Type: documentation | Status: new Priority: normal | Milestone: Component: documentation | Version: Severity: normal | Resolution: Keywords: | ---------------------------+------------------------------------------------ Changes (by des): * type: defect => documentation -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Thu Aug 17 14:29:32 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Thu, 17 Aug 2006 14:29:32 -0000 Subject: [Varnish] #36: setproctitle() on Linux Message-ID: <077.ed084bfa7bd888f3c85a0b1d30381070@projects.linpro.no> #36: setproctitle() on Linux ----------------------+----------------------------------------------------- Reporter: des | Owner: des Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Keywords: ----------------------+----------------------------------------------------- Linux (or rather, glibc) doesn't have setproctitle(); instead, we have a dummy in libcompat. We need a working implementation. We may also want to use our own implementation on *BSD, to get a consistent result across platforms (FreeBSD's setproctitle() is a bit strict in how it ofrmats titles) -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Thu Aug 17 19:39:02 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Thu, 17 Aug 2006 19:39:02 -0000 Subject: [Varnish] #36: setproctitle() on Linux In-Reply-To: <077.ed084bfa7bd888f3c85a0b1d30381070@projects.linpro.no> References: <077.ed084bfa7bd888f3c85a0b1d30381070@projects.linpro.no> Message-ID: <086.2aefecd15f754d487637d09a13169ad5@projects.linpro.no> #36: setproctitle() on Linux ----------------------+----------------------------------------------------- Reporter: des | Owner: des Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Comment (by phk): It is not clear to me how useful or necessary setproctitle() is. The main benefit I see is that we have the ability to distinguish the master and slave processes so people know which one to kill with a venegance. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Fri Aug 18 20:21:12 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Fri, 18 Aug 2006 20:21:12 -0000 Subject: [Varnish] #15: per-client log tailer In-Reply-To: <077.3618b96d691f438f64b61c220162acc5@projects.linpro.no> References: <077.3618b96d691f438f64b61c220162acc5@projects.linpro.no> Message-ID: <086.2673a11a4f63fe300f6f231200ecae9b@projects.linpro.no> #15: per-client log tailer -------------------------+-------------------------------------------------- Reporter: des | Owner: phk Type: enhancement | Status: new Priority: normal | Milestone: Component: varnishlog | Version: trunk Severity: normal | Resolution: Keywords: | -------------------------+-------------------------------------------------- Comment (by phk): This is now somewhat possible in the ordered (-o) mode if the IP range can be described as a regexp. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Fri Aug 18 20:22:25 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Fri, 18 Aug 2006 20:22:25 -0000 Subject: [Varnish] #31: slow client timeout In-Reply-To: <077.0477aaf3a3c8bc6e16fde70f028d88ae@projects.linpro.no> References: <077.0477aaf3a3c8bc6e16fde70f028d88ae@projects.linpro.no> Message-ID: <086.ba821d5d8fe1d9eb0ff5eb4120e11d2d@projects.linpro.no> #31: slow client timeout ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: closed Priority: normal | Milestone: Component: varnishd | Version: trunk Severity: normal | Resolution: fixed Keywords: | ----------------------+----------------------------------------------------- Changes (by phk): * status: new => closed * resolution: => fixed Comment: This has been solved about a week ago. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sat Aug 19 22:08:47 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sat, 19 Aug 2006 22:08:47 -0000 Subject: [Varnish] #37: use SO_RCVTIMEO Message-ID: <077.1b8cb12de86288519cd478b74baf5928@projects.linpro.no> #37: use SO_RCVTIMEO -------------------------+-------------------------------------------------- Reporter: phk | Owner: phk Type: enhancement | Status: new Priority: normal | Milestone: Component: varnishd | Version: Severity: normal | Keywords: -------------------------+-------------------------------------------------- Using SO_RCVTIMEO might simplify the acceptor(s) a bit. Does Linux have it ? -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sat Aug 19 22:10:20 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sat, 19 Aug 2006 22:10:20 -0000 Subject: [Varnish] #38: Fault tree for performance diagnosis Message-ID: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> #38: Fault tree for performance diagnosis ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: Severity: normal | Keywords: ----------------------+----------------------------------------------------- This is a fault tree we can work through to eliminate theories: Lower than expected traffic handling Alteon not allocating traffic Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Alteon not allocating traffic Bad health-check response time Health-check failues TCP/session setup failures or rejections TCP/session count confusion Bad Varnish responsetime Bad health-check response time Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Health-check failues Packet loss Packet delay TCP/session setup failures or rejections Packet loss Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD resource starvation Packet delay Alteon bugs Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD resource starvation TCP/session setup failures or rejections FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD firewalling FreeBSD routing FreeBSD resource starvation Varnish acceptor bugs Varnish acceptor resource starvation Bad Varnish responsetime varnish acceptor bugs varnish response bugs varnish lock contention varnish resource starvation thread library schedule bugs FreeBSD rate limiting FreeBSD resource startvation (sendfile ?) -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sat Aug 19 22:21:30 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sat, 19 Aug 2006 22:21:30 -0000 Subject: [Varnish] #38: Fault tree for performance diagnosis In-Reply-To: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> References: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> Message-ID: <086.432acaba7ebb9e340dfe15e334559045@projects.linpro.no> #38: Fault tree for performance diagnosis ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Old description: > This is a fault tree we can work through to eliminate theories: > > Lower than expected traffic handling > Alteon not allocating traffic > Packet loss > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Alteon not allocating traffic > Bad health-check response time > Health-check failues > TCP/session setup failures or rejections > TCP/session count confusion > Bad Varnish responsetime > > Bad health-check response time > Packet loss > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Health-check failues > Packet loss > Packet delay > TCP/session setup failures or rejections > > Packet loss > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD resource starvation > > Packet delay > Alteon bugs > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD resource starvation > > TCP/session setup failures or rejections > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD firewalling > FreeBSD routing > FreeBSD resource starvation > Varnish acceptor bugs > Varnish acceptor resource starvation > > Bad Varnish responsetime > varnish acceptor bugs > varnish response bugs > varnish lock contention > varnish resource starvation > thread library schedule bugs > FreeBSD rate limiting > FreeBSD resource startvation (sendfile ?) New description: This is a fault tree we can work through to eliminate theories: Lower than expected traffic handling Alteon not allocating traffic Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Alteon not allocating traffic Bad health-check response time Health-check failues TCP/session setup failures or rejections TCP/session count confusion Bad Varnish responsetime Bad health-check response time Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Health-check failues Packet loss Packet delay TCP/session setup failures or rejections Packet loss Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD resource starvation Packet delay Alteon bugs Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD resource starvation TCP/session setup failures or rejections FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD firewalling FreeBSD routing FreeBSD resource starvation Varnish acceptor bugs Varnish acceptor resource starvation Bad Varnish responsetime varnish acceptor bugs varnish response bugs varnish lock contention varnish resource starvation thread library schedule bugs FreeBSD rate limiting FreeBSD resource starvation (sendfile ?) FreeBSD resource starvation is sysctl kern.ipc.somaxconn: 128 enough ? -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sun Aug 20 07:48:10 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sun, 20 Aug 2006 07:48:10 -0000 Subject: [Varnish] #37: use SO_RCVTIMEO In-Reply-To: <077.1b8cb12de86288519cd478b74baf5928@projects.linpro.no> References: <077.1b8cb12de86288519cd478b74baf5928@projects.linpro.no> Message-ID: <086.91c00e868da7a9d231647f2b2b11edd0@projects.linpro.no> #37: use SO_RCVTIMEO -------------------------+-------------------------------------------------- Reporter: phk | Owner: phk Type: enhancement | Status: new Priority: lowest | Milestone: Component: varnishd | Version: Severity: normal | Resolution: Keywords: | -------------------------+-------------------------------------------------- Changes (by phk): * priority: normal => lowest Comment: I did a quick check with the kqueue acceptor on FreeBSD-currnet and that doesn't react to SO_RCVTIMEO (or SO_RCVTIMEO does not happen, I can't tell). It might still help the poll based acceptors though. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sun Aug 20 08:01:26 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sun, 20 Aug 2006 08:01:26 -0000 Subject: [Varnish] #38: Fault tree for performance diagnosis In-Reply-To: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> References: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> Message-ID: <086.cef563a0d3ec841174b67e2fa6f9c287@projects.linpro.no> #38: Fault tree for performance diagnosis ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Old description: > This is a fault tree we can work through to eliminate theories: > > Lower than expected traffic handling > Alteon not allocating traffic > Packet loss > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Alteon not allocating traffic > Bad health-check response time > Health-check failues > TCP/session setup failures or rejections > TCP/session count confusion > Bad Varnish responsetime > > Bad health-check response time > Packet loss > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Health-check failues > Packet loss > Packet delay > TCP/session setup failures or rejections > > Packet loss > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD resource starvation > > Packet delay > Alteon bugs > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD resource starvation > > TCP/session setup failures or rejections > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD firewalling > FreeBSD routing > FreeBSD resource starvation > Varnish acceptor bugs > Varnish acceptor resource starvation > > Bad Varnish responsetime > varnish acceptor bugs > varnish response bugs > varnish lock contention > varnish resource starvation > thread library schedule bugs > FreeBSD rate limiting > FreeBSD resource starvation (sendfile ?) > > FreeBSD resource starvation > is sysctl kern.ipc.somaxconn: 128 enough ? New description: This is a fault tree we can work through to eliminate theories: Lower than expected traffic handling Alteon not allocating traffic ''Packet loss'' Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Alteon not allocating traffic Bad health-check response time Health-check failues TCP/session setup failures or rejections TCP/session count confusion Bad Varnish responsetime Bad health-check response time ''Packet loss'' Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Health-check failues ''Packet loss'' Packet delay TCP/session setup failures or rejections Packet loss Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD resource starvation Packet delay Alteon bugs Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD resource starvation TCP/session setup failures or rejections FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD firewalling FreeBSD routing FreeBSD resource starvation Varnish acceptor bugs Varnish acceptor resource starvation Bad Varnish responsetime varnish acceptor bugs varnish response bugs varnish lock contention varnish resource starvation thread library schedule bugs FreeBSD rate limiting FreeBSD resource starvation (sendfile ?) FreeBSD resource starvation is sysctl kern.ipc.somaxconn: 128 enough ? Packet loss An aggressive ping-test does not show any losses. This is not conclusive, but at least we can defer further Investigation until later. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sun Aug 20 08:02:17 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sun, 20 Aug 2006 08:02:17 -0000 Subject: [Varnish] #38: Fault tree for performance diagnosis In-Reply-To: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> References: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> Message-ID: <086.02e15154703ca41f85a65b3f9b06c45e@projects.linpro.no> #38: Fault tree for performance diagnosis ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Old description: > This is a fault tree we can work through to eliminate theories: > > Lower than expected traffic handling > Alteon not allocating traffic > ''Packet loss'' > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Alteon not allocating traffic > Bad health-check response time > Health-check failues > TCP/session setup failures or rejections > TCP/session count confusion > Bad Varnish responsetime > > Bad health-check response time > ''Packet loss'' > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Health-check failues > ''Packet loss'' > Packet delay > TCP/session setup failures or rejections > > Packet loss > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD resource starvation > > Packet delay > Alteon bugs > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD resource starvation > > TCP/session setup failures or rejections > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD firewalling > FreeBSD routing > FreeBSD resource starvation > Varnish acceptor bugs > Varnish acceptor resource starvation > > Bad Varnish responsetime > varnish acceptor bugs > varnish response bugs > varnish lock contention > varnish resource starvation > thread library schedule bugs > FreeBSD rate limiting > FreeBSD resource starvation (sendfile ?) > > FreeBSD resource starvation > is sysctl kern.ipc.somaxconn: 128 enough ? > > Packet loss > An aggressive ping-test does not show any losses. > This is not conclusive, but at least we can defer further > Investigation until later. New description: This is a fault tree we can work through to eliminate theories: {{{ Lower than expected traffic handling Alteon not allocating traffic ''Packet loss'' Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Alteon not allocating traffic Bad health-check response time Health-check failues TCP/session setup failures or rejections TCP/session count confusion Bad Varnish responsetime Bad health-check response time ''Packet loss'' Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Health-check failues ''Packet loss'' Packet delay TCP/session setup failures or rejections Packet loss Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD resource starvation Packet delay Alteon bugs Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD resource starvation TCP/session setup failures or rejections FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD firewalling FreeBSD routing FreeBSD resource starvation Varnish acceptor bugs Varnish acceptor resource starvation Bad Varnish responsetime varnish acceptor bugs varnish response bugs varnish lock contention varnish resource starvation thread library schedule bugs FreeBSD rate limiting FreeBSD resource starvation (sendfile ?) FreeBSD resource starvation is sysctl kern.ipc.somaxconn: 128 enough ? Packet loss An aggressive ping-test does not show any losses. This is not conclusive, but at least we can defer further Investigation until later. }}} -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sun Aug 20 08:08:52 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sun, 20 Aug 2006 08:08:52 -0000 Subject: [Varnish] #38: Fault tree for performance diagnosis In-Reply-To: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> References: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> Message-ID: <086.da3c37708a4b3696827ad2034bae098a@projects.linpro.no> #38: Fault tree for performance diagnosis ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Old description: > This is a fault tree we can work through to eliminate theories: > > {{{ > Lower than expected traffic handling > Alteon not allocating traffic > ''Packet loss'' > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Alteon not allocating traffic > Bad health-check response time > Health-check failues > TCP/session setup failures or rejections > TCP/session count confusion > Bad Varnish responsetime > > Bad health-check response time > ''Packet loss'' > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Health-check failues > ''Packet loss'' > Packet delay > TCP/session setup failures or rejections > > Packet loss > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD resource starvation > > Packet delay > Alteon bugs > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD resource starvation > > TCP/session setup failures or rejections > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD firewalling > FreeBSD routing > FreeBSD resource starvation > Varnish acceptor bugs > Varnish acceptor resource starvation > > Bad Varnish responsetime > varnish acceptor bugs > varnish response bugs > varnish lock contention > varnish resource starvation > thread library schedule bugs > FreeBSD rate limiting > FreeBSD resource starvation (sendfile ?) > > FreeBSD resource starvation > is sysctl kern.ipc.somaxconn: 128 enough ? > > Packet loss > An aggressive ping-test does not show any losses. > This is not conclusive, but at least we can defer further > Investigation until later. > }}} New description: This is a fault tree we can work through to eliminate theories: {{{ Lower than expected traffic handling Alteon not allocating traffic -Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Alteon not allocating traffic Bad health-check response time Health-check failues TCP/session setup failures or rejections TCP/session count confusion Bad Varnish responsetime Bad health-check response time -Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Health-check failues -Packet loss Packet delay TCP/session setup failures or rejections Packet loss Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD resource starvation Packet delay Alteon bugs Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD resource starvation TCP/session setup failures or rejections FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD firewalling FreeBSD routing FreeBSD resource starvation Varnish acceptor bugs Varnish acceptor resource starvation Bad Varnish responsetime varnish acceptor bugs varnish response bugs varnish lock contention varnish resource starvation thread library schedule bugs FreeBSD rate limiting FreeBSD resource starvation (sendfile ?) FreeBSD resource starvation is sysctl kern.ipc.somaxconn: 128 enough ? Packet loss An aggressive ping-test does not show any losses. This is not conclusive, but at least we can defer further Investigation until later. Packet delay This one is mightlig suspect. The 200msec delays we see against the Alteon is not only present with the alteons health-check but also on a ping: c21# ping -i .001 -c 1000 -q 10.0.2.1 round-trip min/avg/max/stddev = 0.171/0.354/1.038/0.109 ms c21# ping -i .001 -c 1000 -q 10.0.0.2 round-trip min/avg/max/stddev = 0.193/18.655/220.481/46.626 ms But the squids also see the 200msec delay in a ping test: c1# ping -i .001 -c 1000 -q 10.0.0.2 round-trip min/avg/max/stddev = 0.219/22.747/222.899/50.803 ms So this is not unique to us. }}} -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sun Aug 20 08:13:03 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sun, 20 Aug 2006 08:13:03 -0000 Subject: [Varnish] #38: Fault tree for performance diagnosis In-Reply-To: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> References: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> Message-ID: <086.6d948f963d0a16b580cd0630776e195f@projects.linpro.no> #38: Fault tree for performance diagnosis ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Old description: > This is a fault tree we can work through to eliminate theories: > > {{{ > Lower than expected traffic handling > Alteon not allocating traffic > -Packet loss > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Alteon not allocating traffic > Bad health-check response time > Health-check failues > TCP/session setup failures or rejections > TCP/session count confusion > Bad Varnish responsetime > > Bad health-check response time > -Packet loss > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Health-check failues > -Packet loss > Packet delay > TCP/session setup failures or rejections > > Packet loss > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD resource starvation > > Packet delay > Alteon bugs > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD resource starvation > > TCP/session setup failures or rejections > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD firewalling > FreeBSD routing > FreeBSD resource starvation > Varnish acceptor bugs > Varnish acceptor resource starvation > > Bad Varnish responsetime > varnish acceptor bugs > varnish response bugs > varnish lock contention > varnish resource starvation > thread library schedule bugs > FreeBSD rate limiting > FreeBSD resource starvation (sendfile ?) > > FreeBSD resource starvation > is sysctl kern.ipc.somaxconn: 128 enough ? > > Packet loss > An aggressive ping-test does not show any losses. > This is not conclusive, but at least we can defer further > Investigation until later. > > Packet delay > This one is mightlig suspect. > > The 200msec delays we see against the Alteon is not only > present with the alteons health-check but also on a ping: > c21# ping -i .001 -c 1000 -q 10.0.2.1 > round-trip min/avg/max/stddev = 0.171/0.354/1.038/0.109 ms > c21# ping -i .001 -c 1000 -q 10.0.0.2 > round-trip min/avg/max/stddev = 0.193/18.655/220.481/46.626 ms > But the squids also see the 200msec delay in a ping test: > c1# ping -i .001 -c 1000 -q 10.0.0.2 > round-trip min/avg/max/stddev = 0.219/22.747/222.899/50.803 ms > So this is not unique to us. > > }}} New description: This is a fault tree we can work through to eliminate theories: {{{ Lower than expected traffic handling Alteon not allocating traffic -Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Alteon not allocating traffic Bad health-check response time Health-check failues TCP/session setup failures or rejections TCP/session count confusion Bad Varnish responsetime Bad health-check response time -Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Health-check failues -Packet loss Packet delay TCP/session setup failures or rejections Packet loss Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD resource starvation An aggressive ping-test does not show any losses. This is not conclusive, but at least we can defer further Investigation until later. Packet delay Alteon bugs Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD resource starvation This one is mightlig suspect. The 200msec delays we see against the Alteon is not only present with the alteons health-check but also on a ping: c21# ping -i .001 -c 1000 -q 10.0.2.1 round-trip min/avg/max/stddev = 0.171/0.354/1.038/0.109 ms c21# ping -i .001 -c 1000 -q 10.0.0.2 round-trip min/avg/max/stddev = 0.193/18.655/220.481/46.626 ms But the squids also see the 200msec delay in a ping test: c1# ping -i .001 -c 1000 -q 10.0.0.2 round-trip min/avg/max/stddev = 0.219/22.747/222.899/50.803 ms So this is not unique to us. TCP/session setup failures or rejections FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD firewalling FreeBSD routing FreeBSD resource starvation Varnish acceptor bugs Varnish acceptor resource starvation Bad Varnish responsetime varnish acceptor bugs varnish response bugs varnish lock contention varnish resource starvation thread library schedule bugs FreeBSD rate limiting FreeBSD resource starvation (sendfile ?) FreeBSD resource starvation is sysctl kern.ipc.somaxconn: 128 enough ? }}} -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Sun Aug 20 08:16:43 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Sun, 20 Aug 2006 08:16:43 -0000 Subject: [Varnish] #38: Fault tree for performance diagnosis In-Reply-To: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> References: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> Message-ID: <086.4d00212c7f73cada5d0663d815fa821d@projects.linpro.no> #38: Fault tree for performance diagnosis ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: normal | Milestone: Component: varnishd | Version: Severity: normal | Resolution: Keywords: | ----------------------+----------------------------------------------------- Old description: > This is a fault tree we can work through to eliminate theories: > > {{{ > Lower than expected traffic handling > Alteon not allocating traffic > -Packet loss > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Alteon not allocating traffic > Bad health-check response time > Health-check failues > TCP/session setup failures or rejections > TCP/session count confusion > Bad Varnish responsetime > > Bad health-check response time > -Packet loss > Packet delay > TCP/session setup failures or rejections > Bad Varnish responsetime > > Health-check failues > -Packet loss > Packet delay > TCP/session setup failures or rejections > > Packet loss > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD resource starvation > > An aggressive ping-test does not show any losses. > This is not conclusive, but at least we can defer further > Investigation until later. > > Packet delay > Alteon bugs > Alteon interface > GigE switch > bge1 interface > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD resource starvation > > This one is mightlig suspect. > > The 200msec delays we see against the Alteon is not only > present with the alteons health-check but also on a ping: > c21# ping -i .001 -c 1000 -q 10.0.2.1 > round-trip min/avg/max/stddev = 0.171/0.354/1.038/0.109 ms > c21# ping -i .001 -c 1000 -q 10.0.0.2 > round-trip min/avg/max/stddev = 0.193/18.655/220.481/46.626 ms > But the squids also see the 200msec delay in a ping test: > c1# ping -i .001 -c 1000 -q 10.0.0.2 > round-trip min/avg/max/stddev = 0.219/22.747/222.899/50.803 ms > So this is not unique to us. > > > TCP/session setup failures or rejections > FreeBSD Network stack bugs > FreeBSD rate limiting > FreeBSD firewalling > FreeBSD routing > FreeBSD resource starvation > Varnish acceptor bugs > Varnish acceptor resource starvation > > Bad Varnish responsetime > varnish acceptor bugs > varnish response bugs > varnish lock contention > varnish resource starvation > thread library schedule bugs > FreeBSD rate limiting > FreeBSD resource starvation (sendfile ?) > > FreeBSD resource starvation > is sysctl kern.ipc.somaxconn: 128 enough ? > > }}} New description: This is a fault tree we can work through to eliminate theories: {{{ Lower than expected traffic handling Alteon not allocating traffic -Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Alteon not allocating traffic Bad health-check response time Health-check failues TCP/session setup failures or rejections TCP/session count confusion Bad Varnish responsetime When using the "leastconn" metric, the trafic allocation is very sensitive to the varnish sess_timeout setting, the lower I set it, the more traffic I get. From a small sample of data, it looks like there is little benefit from setting it above five seconds anyway. Bad health-check response time -Packet loss Packet delay TCP/session setup failures or rejections Bad Varnish responsetime Health-check failues -Packet loss Packet delay TCP/session setup failures or rejections Packet loss Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD resource starvation An aggressive ping-test does not show any losses. This is not conclusive, but at least we can defer further Investigation until later. Packet delay Alteon bugs Alteon interface GigE switch bge1 interface FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD resource starvation This one is mightlig suspect. The 200msec delays we see against the Alteon is not only present with the alteons health-check but also on a ping: c21# ping -i .001 -c 1000 -q 10.0.2.1 round-trip min/avg/max/stddev = 0.171/0.354/1.038/0.109 ms c21# ping -i .001 -c 1000 -q 10.0.0.2 round-trip min/avg/max/stddev = 0.193/18.655/220.481/46.626 ms But the squids also see the 200msec delay in a ping test: c1# ping -i .001 -c 1000 -q 10.0.0.2 round-trip min/avg/max/stddev = 0.219/22.747/222.899/50.803 ms So this is not unique to us. TCP/session setup failures or rejections FreeBSD Network stack bugs FreeBSD rate limiting FreeBSD firewalling FreeBSD routing FreeBSD resource starvation Varnish acceptor bugs Varnish acceptor resource starvation Bad Varnish responsetime varnish acceptor bugs varnish response bugs varnish lock contention varnish resource starvation thread library schedule bugs FreeBSD rate limiting FreeBSD resource starvation (sendfile ?) FreeBSD resource starvation is sysctl kern.ipc.somaxconn: 128 enough ? }}} -- Ticket URL: Varnish The Varnish HTTP Accelerator From des at linpro.no Mon Aug 21 13:02:55 2006 From: des at linpro.no (Dag-Erling =?iso-8859-1?Q?Sm=F8rgrav?=) Date: Mon, 21 Aug 2006 15:02:55 +0200 Subject: [Varnish] #38: Fault tree for performance diagnosis References: <077.fd93c72b201dddfe334bbfe7abfee800@projects.linpro.no> <086.4d00212c7f73cada5d0663d815fa821d@projects.linpro.no> Message-ID: Varnish writes: > #38: Fault tree for performance diagnosis > [...] Wouldn't it be more useful to have this on a wiki page? DES -- Dag-Erling Sm?rgrav Senior Software Developer Linpro AS - www.linpro.no From phk at phk.freebsd.dk Mon Aug 21 14:19:03 2006 From: phk at phk.freebsd.dk (Poul-Henning Kamp) Date: Mon, 21 Aug 2006 14:19:03 +0000 Subject: [Varnish] #38: Fault tree for performance diagnosis In-Reply-To: Your message of "Mon, 21 Aug 2006 15:02:55 +0200." Message-ID: <17691.1156169943@critter.freebsd.dk> In message , Dag-Erling =?iso-8859-1?Q?Sm=F8rgra v?= writes: >Varnish writes: >> #38: Fault tree for performance diagnosis >> [...] > >Wouldn't it be more useful to have this on a wiki page? Well,it pertaine{s/ed} to one specific bug so I put it in a ticket. I think the problem is solved btw. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk at FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. From varnish-bugs at projects.linpro.no Mon Aug 21 17:13:43 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Mon, 21 Aug 2006 17:13:43 -0000 Subject: [Varnish] #39: workspace confusion Message-ID: <077.1f3840d0ec258e4829efe7081656a45b@projects.linpro.no> #39: workspace confusion ----------------------+----------------------------------------------------- Reporter: phk | Owner: phk Type: defect | Status: new Priority: low | Milestone: Component: varnishd | Version: Severity: normal | Keywords: ----------------------+----------------------------------------------------- There is some confusion about how many HTTP protocol messages go into one workspace. Audit the code so that params->mem_workspace holds one HTTP protocol message. -- Ticket URL: Varnish The Varnish HTTP Accelerator From varnish-bugs at projects.linpro.no Wed Aug 23 13:47:46 2006 From: varnish-bugs at projects.linpro.no (Varnish) Date: Wed, 23 Aug 2006 13:47:46 -0000 Subject: [Varnish] #40: AC_REPLACE_FUNCS Message-ID: <077.8fd5bd6793e970b28ffdc46a5c90ba5b@projects.linpro.no> #40: AC_REPLACE_FUNCS -------------------------+-------------------------------------------------- Reporter: des | Owner: des Type: enhancement | Status: new Priority: normal | Milestone: Varnish 1.0 Component: build | Version: trunk Severity: normal | Keywords: -------------------------+-------------------------------------------------- The current system for selecting which parts of libcompat to build should be replaced with the correct autoconf idiom, AC_REPLACE_FUNCS. -- Ticket URL: Varnish The Varnish HTTP Accelerator