ef8381733bcc14dc6a916629d862d31ad4e70795 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/ff4c830c840fc6f93f3aa9f7a8582a4e4dc3ada4 ff4c830c840fc6f93f3aa9f7a8582a4e4dc3ada4

2009-04-10T23:25:07-07:00

version bump. 6c8dc60fff98ad5625c572596981104282650fc2 dormando dormando@rydia.net b8a39193ea9b62bd0293f345f07dfcd7e942f82f dormando dormando@rydia.net http://github.com/dustin/memcached/commit/ef8381733bcc14dc6a916629d862d31ad4e70795 ef8381733bcc14dc6a916629d862d31ad4e70795

2009-04-10T23:13:50-07:00

remove 'stats maps' command. blah blah something randomized stack reordering. please don't run memcached in any form where arbitrary commands can be executed against it by random unknown users. Apparently you can add to the long list of potential issues, that if someday somewhere sometime someone maybe finds a buffer overflow in memcached, *and* can execute arbitrary stats commands against your server, they could utilize both to form VOLTRON. this *isn't* a critical fix for this next release, but I'm tossing it in for completeness, since someone has reported the bug. 4e2fc9b9abcbd9b1f81d8931eba4c09affd33f3e dormando dormando@rydia.net 9d3bad493922be9561bb206118bac038d956bd0a dormando dormando@rydia.net http://github.com/dustin/memcached/commit/b8a39193ea9b62bd0293f345f07dfcd7e942f82f b8a39193ea9b62bd0293f345f07dfcd7e942f82f

2009-04-10T23:02:14-07:00

fix resumption of accept loop under multithread critical fix: under multithreaded mode, in version 1.2.7, memcached would not resume accepting connections after hitting the max connections limit. this is due to a pre-existing bug in the accept_new_conns code, which was hit when the "fix" was rolled in to ensure listening threads never did real work. Previously, at least one closing connection would (randomly) be on the accepting thread, so the test for (are we the listening thread?) would still work. 66b24c7946e24ee0c7397045865f06a07c128497 dormando dormando@rydia.net c0a382754a1432c10ac939da5970434ac0361d73 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/9d3bad493922be9561bb206118bac038d956bd0a 9d3bad493922be9561bb206118bac038d956bd0a

2009-04-10T21:10:36-07:00

-b now works. a685323785e5ceb46ff73eaa89e4ea1fc6cc5969 dormando dormando@rydia.net 127709a9521c81c6bbda5526dd6316aa5ccb8dd8 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/c0a382754a1432c10ac939da5970434ac0361d73 c0a382754a1432c10ac939da5970434ac0361d73

2009-04-02T18:33:11-07:00

version bump. 38a39046d9ea790b81dce5b35219219d15cd2e60 dormando dormando@rydia.net 55898297e7bdbada33f71664fe264fd70c4ae609 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/127709a9521c81c6bbda5526dd6316aa5ccb8dd8 127709a9521c81c6bbda5526dd6316aa5ccb8dd8

2009-04-01T14:20:38-07:00

'stats reset' resets new main stats. b6ac6331d2e8939fdd2559ea84225ee0120756e8 dormando dormando@rydia.net 98c8b5066f9e1264a309178b268be6dd21405938 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/55898297e7bdbada33f71664fe264fd70c4ae609 55898297e7bdbada33f71664fe264fd70c4ae609

2009-03-30T15:30:45-07:00

throw in notes for the new stats... 9b391afaa394d7537b6c43a01cb5db831333585d dormando dormando@rydia.net d144520c42eff3ae4b2cd1272f41759d2c9d6283 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/98c8b5066f9e1264a309178b268be6dd21405938 98c8b5066f9e1264a309178b268be6dd21405938

2009-03-30T15:25:24-07:00

bump version. 5e024a008b2d32148fb6499f7b895f09b19497c2 dormando dormando@rydia.net c607401efd030d019a47bb100fc4d397801143ce dormando dormando@rydia.net http://github.com/dustin/memcached/commit/d144520c42eff3ae4b2cd1272f41759d2c9d6283 d144520c42eff3ae4b2cd1272f41759d2c9d6283

2009-03-30T15:24:31-07:00

update changelog. 648a07252e62de98e02179c2cdd11c117a2c6497 dormando dormando@rydia.net 8620ef062f0cdacffcdcd6d40c70fcdc9cdb9c02 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/c607401efd030d019a47bb100fc4d397801143ce c607401efd030d019a47bb100fc4d397801143ce

2009-03-28T19:48:45-07:00

2009-03-27T16:12:46-07:00

two new troubleshooting stats accepting_conns for completeness, and listen_disabled_num to see how many times you've hit maxconns and disabled incoming connections. probably a good stat to monitor and flip out on. 446b9a975497d5470b59a2c805d14f22cef925d7 dormando dormando@rydia.net 758f6548acb88d9f446e7369a7564d5382d1e2d2 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/8620ef062f0cdacffcdcd6d40c70fcdc9cdb9c02 8620ef062f0cdacffcdcd6d40c70fcdc9cdb9c02

2009-03-28T19:48:39-07:00

2009-03-25T00:55:19-07:00

add a cmd_flush stat shouldn't add much lock contention for just this. I want to add this one stat (mayb a few more?) since it's happened more than once that folks think memcached is broken when a cron or something is calling 'flush_all' once a minute. 7d232e6c0f25200604ac206e9787df046e5014a9 dormando dormando@rydia.net 7208606e817dc401bcd3fde17038a3dfad604cab dormando dormando@rydia.net http://github.com/dustin/memcached/commit/758f6548acb88d9f446e7369a7564d5382d1e2d2 758f6548acb88d9f446e7369a7564d5382d1e2d2

2009-03-28T19:48:35-07:00

2009-03-24T23:55:09-07:00

print why a key was not found when extra verbose simple logs for simple people. Patch inspired by a bug hunting session with evan weaver. It's been useful a few times since. b7d5fe3da635ecb63c64f4c8ced8b035156d4557 dormando dormando@rydia.net bc3b2e295ef770b682e90a12e2b5e9acfdf85971 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/7208606e817dc401bcd3fde17038a3dfad604cab 7208606e817dc401bcd3fde17038a3dfad604cab

2009-03-28T19:48:13-07:00

2009-03-28T00:16:38-07:00

dumb hack to self-repair stuck slabs since 1.2.6, most of the refcount leaks have been quashed. I still get them in production, extremely rarely. It's possibly we'll have refcount leaks on and off even in the future. This hack acknowledges this and exists since we want to guarantee, as much as possible, that memcached is a stable service. Having to monitor for and restart the service on account of "rare bugs" isn't acceptable. c74064fbf903a46a47153ffd6634cfb5798a7667 dormando dormando@rydia.net cd9f5e17a5db997af16a8fccb7a59efb0c8e52c0 dormando dormando@rydia.net http://github.com/dustin/memcached/commit/bc3b2e295ef770b682e90a12e2b5e9acfdf85971 bc3b2e295ef770b682e90a12e2b5e9acfdf85971

2009-03-28T19:47:42-07:00

2009-03-26T00:19:00-07:00

fix a handful of socket listen bugs. AF_UNSPEC is still necessary for UDP sometimes. We guarantee that at least one address returned from getaddrinfo binds successfully, and in cases of lacking network or ipv6 addresses some of those socket() calls might fail. That's normal. We were bailing on them. This change also removes the need to pass AI_ADDRCONFIG on machines with ipv6 stacks disabled. 627a3672faf3762199bab7b3d9aa6e734c5db84f dormando dormando@rydia.net bf832f3ae339ce8dbcd72dc4d38eb2985a122a32 Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/cd9f5e17a5db997af16a8fccb7a59efb0c8e52c0 cd9f5e17a5db997af16a8fccb7a59efb0c8e52c0

2009-03-28T19:45:58-07:00

2009-03-17T15:02:07-07:00

stats slab's used_chunks should show chunks put to use It was a bit unclear what it was doing before, but it started out with a value equal to total_chunks, which was surely wrong. This change and its accompanying test ensure the value makes a bit more sense. a6bf0f69614179f3be4c5bad00fba7c034b1e20f dormando dormando@rydia.net 34a40c7d928df845c89fd6d05fd6f61bfac3cffa Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/bf832f3ae339ce8dbcd72dc4d38eb2985a122a32 bf832f3ae339ce8dbcd72dc4d38eb2985a122a32

2009-03-28T19:45:50-07:00

2009-03-13T20:24:14-07:00

A bit more space for server_stats, and an assertion (bug 27). I couldn't figure out how to get the stats output big enough to exceed 1024, but I accept it might. I've given it a bit more space here and added an assertion to detect when we fail in case we can figure out how to actually have this occur. Maxing out pretty much everything got me up to 828 bytes: STAT pid 35893 STAT uptime 152 STAT time 1237000260 STAT version 1.3.2 STAT pointer_size 32 STAT rusage_user 0.002487 STAT rusage_system 0.005412 STAT curr_connections 4 STAT total_connections 5 STAT connection_structures 5 STAT cmd_get 18446744073709551610 STAT cmd_set 18446744073709551610 STAT get_hits 18446744073709551610 STAT get_misses 18446744073709551610 STAT delete_misses 18446744073709551610 STAT delete_hits 18446744073709551610 STAT incr_misses 18446744073709551610 STAT incr_hits 18446744073709551610 STAT decr_misses 18446744073709551610 STAT decr_hits 18446744073709551610 STAT cas_misses 18446744073709551610 STAT cas_hits 18446744073709551610 STAT cas_badval 18446744073709551610 STAT bytes_read 1 STAT bytes_written 18446744073709551610 STAT limit_maxbytes 67108864 STAT threads 5 STAT bytes 18446744073709551610 STAT curr_items 4294967290 STAT total_items 4294967290 STAT evictions 18446744073709551610 ... assertion removed for the 1.2 tree. 2f4a4f2e79a04fa91ac113c5a80cffa2d555f74a dormando dormando@rydia.net e703ce4cb0b0ee8034ba9c2331877a97ffb6d11a Chris Goffinet cg@chrisgoffinet.com http://github.com/dustin/memcached/commit/34a40c7d928df845c89fd6d05fd6f61bfac3cffa 34a40c7d928df845c89fd6d05fd6f61bfac3cffa

2009-03-28T19:45:44-07:00

2009-03-15T23:08:46-07:00

Added support for setting backlog queue on command line 29f3664bcacc337295af9eec1c86078aa1282a3f dormando dormando@rydia.net 5a44468708aa09fcec469557258a5d69a88811ca Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/e703ce4cb0b0ee8034ba9c2331877a97ffb6d11a e703ce4cb0b0ee8034ba9c2331877a97ffb6d11a

2009-03-12T16:31:29-07:00

2009-01-19T15:45:38-08:00

Build fixes for ubuntu 8.10/64. gcc seems to have been wise to all of the syscalls being cast to void to avoid doing real result checking in unlikely errors. 983c362bc49dc4663f9bfc39b7c966065d467cf2 Dustin Sallings dustin@spy.net 01f377b8c730a02f8efeda63d4dac1803e950fde Evan Klitzke evan@eklitzke.org http://github.com/dustin/memcached/commit/5a44468708aa09fcec469557258a5d69a88811ca 5a44468708aa09fcec469557258a5d69a88811ca

2009-03-12T16:31:29-07:00

2009-03-01T17:18:03-08:00

Use AI_ADDRCONFIG more selectively. This makes running on Linux when disconnected possible. Tests still fail when using the -l option to specify a binding address while offline. This is probably quite unlikely if it weren't for the first test. 3a3ab5905968b1b8671aeba720a8d16452cdaddc Dustin Sallings dustin@spy.net 6973dd54f00aede1ee859f5f312809c80a7b61ac Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/01f377b8c730a02f8efeda63d4dac1803e950fde 01f377b8c730a02f8efeda63d4dac1803e950fde

2009-03-12T16:31:29-07:00

2009-03-11T12:48:46-07:00

Slow down the add test a bit. Solaris with dtrace enabled was rather consistently failing this. Expiration times around 1 can get dropped due to memcached clock resolution. b4d534e172bc69af701e0b1e969899195f16bcb0 Dustin Sallings dustin@spy.net a293b69334c4f9fdfe25f8a7464fc0dfc04ee773 Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/6973dd54f00aede1ee859f5f312809c80a7b61ac 6973dd54f00aede1ee859f5f312809c80a7b61ac

2009-03-12T16:31:29-07:00

2009-03-11T11:31:17-07:00

Shrink lines in protocol.txt to fit within 80 columns. Also wrote a new test that will ensure that that never happens again. 35ebfc2d4a7b86541e94db9a12d686ee835e4708 Dustin Sallings dustin@spy.net e10a96345fd1f17287bbfbcb992a0bd4ab491c1e Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/a293b69334c4f9fdfe25f8a7464fc0dfc04ee773 a293b69334c4f9fdfe25f8a7464fc0dfc04ee773

2009-03-12T16:31:29-07:00

2009-03-11T00:18:57-07:00

Added an add behavior test. Someone in IRC reported this not working one day. Likely a failure on the client side, but it's good to ensure it for us in a test case. 725d8c3b77111a3cd97871c357e07e615f2945dc Dustin Sallings dustin@spy.net 19856aa075ed804a251e67fb5bacf0e0fd63e2c5 Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/e10a96345fd1f17287bbfbcb992a0bd4ab491c1e e10a96345fd1f17287bbfbcb992a0bd4ab491c1e

2009-03-12T16:31:29-07:00

2009-03-01T10:46:09-08:00

Isolate the first server in startup test so it dies sooner (bug 20). Apparently on some version of Linux I don't have, this causes the test to hang for a while as gcov attempts to compute and write out the results from two coverage runs concurrently. It's best to clean up things as quickly as possible in the general case, so this should be done regardless. 732563b9716755be05cc5a2a8bfba83192b1361a Dustin Sallings dustin@spy.net ef5fe8702a46f41176c9b96a7d583877d8a3ab63 Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/19856aa075ed804a251e67fb5bacf0e0fd63e2c5 19856aa075ed804a251e67fb5bacf0e0fd63e2c5

2009-03-12T16:31:29-07:00

2009-02-26T08:49:32-08:00

Fix for bug21. Some strange signness handling around decr caused values that would be signed 64-bit numbers to get dropped to zero. 9a4777f7bdce004499317da192c939261696cb6c Dustin Sallings dustin@spy.net 25eb0c170e19da0d926abfa842e2181056eb929f Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/ef5fe8702a46f41176c9b96a7d583877d8a3ab63 ef5fe8702a46f41176c9b96a7d583877d8a3ab63

2009-03-12T16:31:29-07:00

2009-02-12T18:36:01-08:00

Test cachedump on text protocol. f78f3928eed8b56fd41c5cfd6efc7876c50a87f9 Dustin Sallings dustin@spy.net 3787039e257f7639657e02bb7638554f68353592 Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/25eb0c170e19da0d926abfa842e2181056eb929f 25eb0c170e19da0d926abfa842e2181056eb929f

2009-03-12T16:31:29-07:00

2009-02-11T18:03:14-08:00

Update CAS on non-replace incr/decr. This fixes a problem reported as bug 15 where incr and decr do not change CAS values when they aren't completely replacing the item (which is the typical case). http://code.google.com/p/memcached/issues/detail?id=15 c361eabd93e5bc2386e4a57da15c7e6fb28a4325 Dustin Sallings dustin@spy.net 3f2c0fd1d8e95f371d7f9240ba62750c18457c8c Dustin Sallings dustin@spy.net http://github.com/dustin/memcached/commit/3787039e257f7639657e02bb7638554f68353592 3787039e257f7639657e02bb7638554f68353592

2009-03-12T16:31:29-07:00

2009-03-12T16:30:58-07:00

Ensure there are always N+1 threads so udp dispatch works. If compiled with threads disabled, there would be too few threads to run udp and bad stuff would happen. cabf1c58a1d49ecd5be26e0b3d6c112b5bf80156 Dustin Sallings dustin@spy.net 469915d9dc8615ec33126b3452f758a19825494c dormando dormando@rydia.net http://github.com/dustin/memcached/commit/3f2c0fd1d8e95f371d7f9240ba62750c18457c8c 3f2c0fd1d8e95f371d7f9240ba62750c18457c8c

2008-12-22T12:35:55-08:00

2008-09-06T16:02:07-07:00

Update doc/protocol.txt for the new evicted_time stat. ceeeede39119f02ce5ae96c0ba86d068507d1911 Dustin Sallings dustin@spy.net 28d27b22c1c57906cb3cded87e44a4cf28866bca dormando dormando@rydia.net http://github.com/dustin/memcached/commit/469915d9dc8615ec33126b3452f758a19825494c 469915d9dc8615ec33126b3452f758a19825494c

2008-12-22T12:35:55-08:00

2008-09-06T15:52:08-07:00

Update changelog, common contributors. I'm missing some contributors now. Have to look up your address probably. Please bitch if you've tossed more than a couple small patches and want to be included. 6ada9b8827f1a81f71ca076df322f553a0735c4b Dustin Sallings dustin@spy.net 04319dddabaa06d15407ab6f793b160d3b1c5edb dormando dormando@rydia.net http://github.com/dustin/memcached/commit/28d27b22c1c57906cb3cded87e44a4cf28866bca 28d27b22c1c57906cb3cded87e44a4cf28866bca

2008-12-22T12:35:55-08:00

2008-09-06T15:35:21-07:00

Display final errno status on the various listen failures. "failed to listen on TCP port 11211" is nice, but "Address already in use" is more nice. I have a feeling this isn't 100% correct given the ipv6 code, but either it's good enough or someone else will have to outpatch me. c77be5de454a339eb3c2ca14e3dc3b65390c05ab Dustin Sallings dustin@spy.net