1 2017-03-06 01:11:06	0|gmaxwell|streams.h:407:44: warning: declaration of ‘data’ shadows a member of 'this' [-Wshadow]
  2 2017-03-06 01:11:21	0|gmaxwell|I boggle at shadow warnings being so compiler specific.
  3 2017-03-06 01:11:33	0|gmaxwell|This isn't the case for plain C.
  4 2017-03-06 06:49:37	0|wumpus|yes for C it seems much better defined
  5 2017-03-06 06:51:51	0|wumpus|though it may be historical coincidence that clang and gcc match up so well with most warnings. MSVC on the other hand...
  6 2017-03-06 07:14:42	0|sipa|it hurts.
  7 2017-03-06 07:16:10	0|wumpus|that said, it will probably hurt for any moderately-sized C++ project
  8 2017-03-06 07:17:52	0|wumpus|except for the compilers themselves maybe; I'd (naively) expect those to compile cleanly in themselves
  9 2017-03-06 08:40:42	0|bitcoin-git|13bitcoin/06master 14343ba8f 15practicalswift: [wallet] Remove redundant initialization...
 10 2017-03-06 08:40:42	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/01b7cda91684...56ab672b59da
 11 2017-03-06 08:40:43	0|bitcoin-git|13bitcoin/06master 1456ab672 15Wladimir J. van der Laan: Merge #9576: [wallet] Remove redundant initialization...
 12 2017-03-06 08:40:52	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9576: [wallet] Remove redundant initialization (06master...06remove-redundant-initialization-ii) 02https://github.com/bitcoin/bitcoin/pull/9576
 13 2017-03-06 09:02:26	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/56ab672b59da...5a6af3172254
 14 2017-03-06 09:02:27	0|bitcoin-git|13bitcoin/06master 14e57a1fd 15Russell Yanofsky: Define 7200 second timestamp window constant
 15 2017-03-06 09:02:28	0|bitcoin-git|13bitcoin/06master 145a6af31 15Wladimir J. van der Laan: Merge #9908: Define 7200 second timestamp window constant...
 16 2017-03-06 09:02:47	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9908: Define 7200 second timestamp window constant (06master...06pr/timewin) 02https://github.com/bitcoin/bitcoin/pull/9908
 17 2017-03-06 09:04:50	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/5a6af3172254...48c3429c50fb
 18 2017-03-06 09:04:51	0|bitcoin-git|13bitcoin/06master 14025dec0 15NicolasDorier: [qa] assert_start_raises_init_error
 19 2017-03-06 09:04:51	0|bitcoin-git|13bitcoin/06master 1448c3429 15Wladimir J. van der Laan: Merge #9832: [qa] assert_start_raises_init_error...
 20 2017-03-06 09:05:12	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9832: [qa] assert_start_raises_init_error (06master...06assert_start_raises_init_error) 02https://github.com/bitcoin/bitcoin/pull/9832
 21 2017-03-06 09:06:52	0|bitcoin-git|[13bitcoin] 15laanwj pushed 3 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/48c3429c50fb...9d5fcbfb0889
 22 2017-03-06 09:06:53	0|bitcoin-git|13bitcoin/06master 1499c0e81 15John Newbery: Fix BIP68 activation test
 23 2017-03-06 09:06:53	0|bitcoin-git|13bitcoin/06master 14f5aba8a 15John Newbery: Move tx version 2 standardness check to after bip68 activation
 24 2017-03-06 09:06:54	0|bitcoin-git|13bitcoin/06master 149d5fcbf 15Wladimir J. van der Laan: Merge #9739: Fix BIP68 activation test...
 25 2017-03-06 09:07:10	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9739: Fix BIP68 activation test (06master...06fixbip68testing) 02https://github.com/bitcoin/bitcoin/pull/9739
 26 2017-03-06 09:08:35	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/9d5fcbfb0889...d32581cc29d1
 27 2017-03-06 09:08:36	0|bitcoin-git|13bitcoin/06master 14d32581c 15Wladimir J. van der Laan: Merge #9547: bench: Assert that division by zero is unreachable...
 28 2017-03-06 09:08:36	0|bitcoin-git|13bitcoin/06master 14db07f91 15practicalswift: Assert that what might look like a possible division by zero is actually unreachable
 29 2017-03-06 09:08:50	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9547: bench: Assert that division by zero is unreachable (06master...06avoid-potential-division-by-zero-in-benchmark-state-keeprunning) 02https://github.com/bitcoin/bitcoin/pull/9547
 30 2017-03-06 09:18:40	0|bitcoin-git|[13bitcoin] 15laanwj pushed 4 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/d32581cc29d1...fa625b078b01
 31 2017-03-06 09:18:41	0|bitcoin-git|13bitcoin/06master 14654e044 15Russell Yanofsky: [trivial] Add comment documenting CWalletTx::mapValue
 32 2017-03-06 09:18:41	0|bitcoin-git|13bitcoin/06master 14a1fe944 15Russell Yanofsky: Remove reference to nonexistent "version" wallet transaction mapvalue field...
 33 2017-03-06 09:18:42	0|bitcoin-git|13bitcoin/06master 1487ed396 15Russell Yanofsky: [trivial] Add comment documenting bumpfee mapValues
 34 2017-03-06 09:18:49	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9333: Document CWalletTx::mapValue entries and remove erase of nonexistent "version" entry. (06master...06pr/comment-mapvalue) 02https://github.com/bitcoin/bitcoin/pull/9333
 35 2017-03-06 09:37:46	0|bitcoin-git|[13bitcoin] 15kobake opened pull request #9926: Rename argument name 'data' to prevent shadowing. (06master...06fix-shadow-warning) 02https://github.com/bitcoin/bitcoin/pull/9926
 36 2017-03-06 09:48:07	0|bitcoin-git|[13bitcoin] 15kobake closed pull request #9926: Rename argument name 'data' to prevent shadowing. (06master...06fix-shadow-warning) 02https://github.com/bitcoin/bitcoin/pull/9926
 37 2017-03-06 09:59:06	0|bitcoin-git|[13bitcoin] 15jonasschnelli closed pull request #9256: Fix more CWallet/CWalletDB layer violations (06master...062016/12/ref_walletdb) 02https://github.com/bitcoin/bitcoin/pull/9256
 38 2017-03-06 10:06:27	0|wumpus|jonasschnelli: too bade about 9256 :( might pick it up at some point, need to be able to replace the database backend for the cloudabi port
 39 2017-03-06 10:07:09	0|jonasschnelli|wumpus: Oh. Okay.. I'm currently working on #8574 (which is also a BDB abstraction).
 40 2017-03-06 10:07:10	0|gribble|https://github.com/bitcoin/bitcoin/issues/8574 | [Wallet] refactor CWallet/CWalletDB/CDB by jonasschnelli · Pull Request #8574 · bitcoin/bitcoin · GitHub
 41 2017-03-06 10:07:23	0|jonasschnelli|9256 can be done afterwards..
 42 2017-03-06 10:07:45	0|wumpus|I've noticed that there is a lot of overlap between dbwrapper.h and walletdb.h
 43 2017-03-06 10:08:04	0|wumpus|yeah
 44 2017-03-06 10:08:09	0|jonasschnelli|Yes. We need clean layering first to support multiple db backends
 45 2017-03-06 10:08:41	0|wumpus|I really like the dbwrapper.h. With that I've mangaed to replace the db backend multiple times. Implementing e.g. a walletdb.h for leveldb seemsto be more work
 46 2017-03-06 10:09:01	0|wumpus|(as it doesn't wrap some of the objects)
 47 2017-03-06 10:10:40	0|wumpus|then again - a  pure key/value store may not be what we eventually want for the wallet at all, maybe something like sqlite with indexes is better. But on the short term it'd be useful to switch database backends and just consider key/value stores as a given, abstracted.
 48 2017-03-06 10:10:50	0|jonasschnelli|Yes. Something like that for the wallet db would be great. I first though CDBEnv follows that approach... but not really
 49 2017-03-06 10:11:22	0|wumpus|I think they evolved from the same abstraction, but dbwrapper was updated time and time again to be a better abstraction, while walletdb did not :)
 50 2017-03-06 10:11:41	0|jonasschnelli|Sqlite could be useful. I though about it.... but not sure if we want private keys in there.. also not sure about the compatibility between different sqlite implementations..
 51 2017-03-06 10:12:51	0|jonasschnelli|I still like sipa logdb approach (full in-mem mapping, no on-disk database queries, simple append only log with on-going hash)
 52 2017-03-06 10:12:54	0|wumpus|sqlite actually encourages usage as an applicatiion data format in their faq
 53 2017-03-06 10:13:33	0|wumpus|so I *think* that means they make a commitment to on-disk compatibility. Also their databases are single files, unlike leveldb.
 54 2017-03-06 10:14:15	0|wumpus|yes sure I like that too
 55 2017-03-06 10:14:44	0|wumpus|in any case, for everything we need to improve the abstractions involved
 56 2017-03-06 10:14:54	0|jonasschnelli|For pure transactions and pubkey, I would probably prefere sqlite
 57 2017-03-06 10:15:17	0|wumpus|right - and with HD wallets we need a lot less key storage
 58 2017-03-06 10:15:18	0|jonasschnelli|Yes. Lets focus on abstraction and discuss the new persistence layer later.
 59 2017-03-06 10:15:32	0|wumpus|+private
 60 2017-03-06 10:16:06	0|sipa|the abstractions are complicated by the fact that BDB needs some things (flushing, locking, environment, ...) that nothing else needs
 61 2017-03-06 10:17:04	0|jonasschnelli|Yes. The abstraction needs to support that (at least for the next years)
 62 2017-03-06 10:17:44	0|sipa|no, only until we get rid of BDB
 63 2017-03-06 10:17:53	0|sipa|i hope that's less than a few years
 64 2017-03-06 10:18:09	0|MarcoFalke|Anyone want to review #9880?
 65 2017-03-06 10:18:11	0|gribble|https://github.com/bitcoin/bitcoin/issues/9880 | Verify Tree-SHA512s in merge commits, enforce sigs are not SHA1 by TheBlueMatt · Pull Request #9880 · bitcoin/bitcoin · GitHub
 66 2017-03-06 10:18:23	0|jonasschnelli|hmm.. if we offer a good migration option,... but the migration option needs BDB support... :)
 67 2017-03-06 10:18:29	0|MarcoFalke|Currently only wumpus can merge on master
 68 2017-03-06 10:18:41	0|sipa|jonasschnelli: that can just be a trivial "iterate the keys, convert" - not a full db abstraction
 69 2017-03-06 10:18:51	0|jonasschnelli|sipa: Indeed
 70 2017-03-06 10:34:35	0|jonasschnelli|wumpus, sipa: if you care for wallet db abstraction, this (#8574) could be reviewed:
 71 2017-03-06 10:34:37	0|gribble|https://github.com/bitcoin/bitcoin/issues/8574 | [Wallet] refactor CWallet/CWalletDB/CDB by jonasschnelli · Pull Request #8574 · bitcoin/bitcoin · GitHub
 72 2017-03-06 10:44:10	0|jonasschnelli|MarcoFalke: Oh. Yes. My merged broke verify-commits (and therefor travis on master): https://travis-ci.org/bitcoin/bitcoin/jobs/206917518#L513
 73 2017-03-06 11:00:49	0|wumpus|jonasschnelli: I'll take a look. On the short run I just need something that can replace berkeleydb with a different k/v store, no need for any backward compatiblity or migration
 74 2017-03-06 11:01:12	0|jonasschnelli|Thanks. Yes. That's a different path then...
 75 2017-03-06 11:04:06	0|wumpus|jonasschnelli: not using a database or a custom database would be fine too, but that seems like a lot more work :)
 76 2017-03-06 11:04:54	0|wumpus|and would require a huge patch set to maintain on top
 77 2017-03-06 11:06:31	0|wumpus|which would need to be updated for every upstream wallet change
 78 2017-03-06 11:06:39	0|wumpus|so yeah, switching the k/v store is probably the best option
 79 2017-03-06 11:13:20	0|Victorsueca|I just had a question this morning while at was at shower... Is it safe to defrag while bitcoin core is running?
 80 2017-03-06 11:14:20	0|wumpus|defrag? is that still a thing?
 81 2017-03-06 11:14:44	0|BirneGetreide_|Just don't defrag, ever. It is a waste of time
 82 2017-03-06 11:15:28	0|jonasschnelli|Oh... verify-commits.sh gave a "Segmentation fault: 11"! hah
 83 2017-03-06 11:16:03	0|jonasschnelli|I hope it's not coming from gnupg
 84 2017-03-06 11:17:42	0|wumpus|but no, we can't guarantee that that is safe. Moving filesystem blocks in the background could be safe or unsafe depending on whether the OS keeps track of things properly.
 85 2017-03-06 11:19:34	0|Victorsueca|it's windows so I doubt it does lol
 86 2017-03-06 11:29:26	0|wumpus|right, you didn't even need to ask
 87 2017-03-06 11:43:42	0|jonasschnelli|Hmm... I think the verify-commit.sh segfault 11 im getting is from /bin/sh
 88 2017-03-06 11:44:11	0|jonasschnelli|It happens when the script calls IS_SIGNED() (recursive call)
 89 2017-03-06 11:44:30	0|jonasschnelli|Happens on master was well...
 90 2017-03-06 11:48:17	0|wumpus|ugh :/
 91 2017-03-06 11:50:16	0|jonasschnelli|Can't even attach lldb to /bin/sh due to OSX's "integrity protection"... copied out sh and running now in lldb
 92 2017-03-06 11:51:22	0|sipa|too many recursion levels in the shell?
 93 2017-03-06 11:52:09	0|jonasschnelli|sipa: I think so.
 94 2017-03-06 11:52:13	0|jonasschnelli|Stack trace: http://bitcoin.jonasschnelli.ch/st.txt
 95 2017-03-06 11:53:51	0|jonasschnelli|now running in bash
 96 2017-03-06 11:55:00	0|jonasschnelli|same
 97 2017-03-06 11:57:52	0|jonasschnelli|I think some Ubuntu verions are also affected. http://lists.gnu.org/archive/html/bug-bash/2003-12/msg00007.html
 98 2017-03-06 11:58:58	0|jonasschnelli|Would it make sense to port verify-commits.sh to python? We already rely on python for the rest, right? ping BlueMatt
 99 2017-03-06 11:59:58	0|jonasschnelli|BlueMatt: it seems that the current amount of recursive calls on IS_SIGNED() seems to break some shell implementations (OSX 10.11 and I think also Ubuntu 14.04)
100 2017-03-06 12:06:07	0|wumpus|can we somehow make verify-commits.sh simpler, at least temporary?
101 2017-03-06 12:06:35	0|wumpus|e.g. do the most important check that the HEAD commit is signed, but no more, that shouldn't take any recursion
102 2017-03-06 12:06:54	0|wumpus|then after the script is fixed we can go back to the more comphrehensive one
103 2017-03-06 12:19:06	0|jonasschnelli|wumpus: Yes. That's probably a good idea... maybe we add an arg and by default, it just checks the head
104 2017-03-06 12:38:27	0|bitcoin-git|[13bitcoin] 15jonasschnelli opened pull request #9928: Allow verify-commit.sh to just verify the HEAD commit (Use non-recursive verification by default) (06master...062017/03/vc_simple) 02https://github.com/bitcoin/bitcoin/pull/9928
105 2017-03-06 13:26:33	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/fa625b078b01...8a3b07529d9c
106 2017-03-06 13:26:34	0|bitcoin-git|13bitcoin/06master 147184e25 15Jonas Schnelli: [Wallet] refactor CWallet/CWalletDB/CDB...
107 2017-03-06 13:26:34	0|bitcoin-git|13bitcoin/06master 148a3b075 15Wladimir J. van der Laan: Merge #8574: [Wallet] refactor CWallet/CWalletDB/CDB...
108 2017-03-06 13:26:44	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #8574: [Wallet] refactor CWallet/CWalletDB/CDB (06master...062016/08/bdb_abstraction_2) 02https://github.com/bitcoin/bitcoin/pull/8574
109 2017-03-06 14:13:48	0|wumpus|going to have a stab at making the RPC tests run (optionally) through UNIX sockets
110 2017-03-06 14:14:18	0|wumpus|don't think much is needed besides abstracting 'url' into 'RPC connect info'
111 2017-03-06 14:29:32	0|jonasschnelli|wumpus: haven't thought about the RPC test... great idea.
112 2017-03-06 14:44:27	0|wumpus|jonasschnelli: yep it provides an example on how to use RPC over UNIX sockets in python at the same time
113 2017-03-06 14:55:39	0|bitcoin-git|[13bitcoin] 15laanwj opened pull request #9929: tests: Delete unused function _rpchost_to_args (06master...062017_03_tests_unused) 02https://github.com/bitcoin/bitcoin/pull/9929
114 2017-03-06 15:29:26	0|BlueMatt|jonasschnelli: if you want to feel free, but I wont be able to maintain it anymore :P
115 2017-03-06 15:34:19	0|wumpus|many of us have the same problem with shellscript :p
116 2017-03-06 15:34:44	0|wumpus|but I don't care what language it is in, it just needs to work
117 2017-03-06 15:38:22	0|wumpus|nesting recursion too deep is a problem in any language
118 2017-03-06 15:39:51	0|BlueMatt|yea, i should have removed the recursion when i made it not ever look at the second parent
119 2017-03-06 15:39:55	0|BlueMatt|should be trivial to do
120 2017-03-06 15:42:37	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9924: [UI Styling] Left-Align Tab Bar - optionsdialog.ui (06master...06patch-1) 02https://github.com/bitcoin/bitcoin/pull/9924
121 2017-03-06 15:42:57	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9925: [UI Styling] Left-Align Tab Bar - debugwindow.ui (06master...06patch-2) 02https://github.com/bitcoin/bitcoin/pull/9925
122 2017-03-06 16:20:19	0|bitcoin-git|[13bitcoin] 15laanwj pushed 6 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/8a3b07529d9c...4df8213b98d3
123 2017-03-06 16:20:20	0|bitcoin-git|13bitcoin/06master 14be908a6 15Matt Corallo: Fail merge if there are any symlinks
124 2017-03-06 16:20:20	0|bitcoin-git|13bitcoin/06master 14d9c450f 15Matt Corallo: Verify Tree-SHA512s in merge commits, enforce sigs are not SHA1
125 2017-03-06 16:20:21	0|bitcoin-git|13bitcoin/06master 14eddc77a 15Peter Todd: Add comment re: why SHA1 is disabled
126 2017-03-06 16:20:38	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9880: Verify Tree-SHA512s in merge commits, enforce sigs are not SHA1 (06master...062017-02-validate-sha512) 02https://github.com/bitcoin/bitcoin/pull/9880
127 2017-03-06 16:30:06	0|BlueMatt|god damn it, how did it fail on head
128 2017-03-06 16:48:33	0|BlueMatt|grrrr it doesnt fail locally
129 2017-03-06 16:58:26	0|bitcoin-git|[13bitcoin] 15matthias-g opened pull request #9930: Trivial: Correct indentation (06master...06fix-indent) 02https://github.com/bitcoin/bitcoin/pull/9930
130 2017-03-06 17:06:30	0|bitcoin-git|13bitcoin/06master 14b23dcd2 15John Newbery: Fix segwit getblocktemplate test.
131 2017-03-06 17:06:30	0|bitcoin-git|13bitcoin/06master 14c78adbf 15Wladimir J. van der Laan: Merge #9843: Fix segwit getblocktemplate test...
132 2017-03-06 17:06:30	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/4df8213b98d3...c78adbf450be
133 2017-03-06 17:06:51	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9843: Fix segwit getblocktemplate test (06master...06fixsegwitgetblocktemplate) 02https://github.com/bitcoin/bitcoin/pull/9843
134 2017-03-06 17:33:01	0|bitcoin-git|13bitcoin/06master 1499fecf8 15Wladimir J. van der Laan: tests: Delete unused function _rpchost_to_args...
135 2017-03-06 17:33:01	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/c78adbf450be...d5ce14e22338
136 2017-03-06 17:33:02	0|bitcoin-git|13bitcoin/06master 14d5ce14e 15Wladimir J. van der Laan: Merge #9929: tests: Delete unused function _rpchost_to_args...
137 2017-03-06 17:33:21	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9929: tests: Delete unused function _rpchost_to_args (06master...062017_03_tests_unused) 02https://github.com/bitcoin/bitcoin/pull/9929
138 2017-03-06 17:33:46	0|bitcoin-git|13bitcoin/06master 14188f89c 15Gregory Sanders: Disallow copy of CReserveKeys
139 2017-03-06 17:33:46	0|bitcoin-git|13bitcoin/06master 1472fb515 15Wladimir J. van der Laan: Merge #9906: Disallow copy constructor CReserveKeys...
140 2017-03-06 17:33:46	0|bitcoin-git|[13bitcoin] 15laanwj pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/d5ce14e22338...72fb5158b1c8
141 2017-03-06 17:34:11	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9906: Disallow copy constructor CReserveKeys (06master...06noreservecopy) 02https://github.com/bitcoin/bitcoin/pull/9906
142 2017-03-06 17:36:17	0|morcos|Call for another set of eyes on #9602..   I think it has enough technical review now, but couldn't hurt to have some more people review the open questions in the PR message and make sure they are ok with the changes made.
143 2017-03-06 17:36:20	0|gribble|https://github.com/bitcoin/bitcoin/issues/9602 | Remove coin age priority and free transactions - implementation by morcos · Pull Request #9602 · bitcoin/bitcoin · GitHub
144 2017-03-06 17:36:27	0|morcos|And then lets get it merged please
145 2017-03-06 17:54:22	0|gmaxwell|wumpus: for the unix domain socket bitcoin-cli, could you have the client behind an actual feature test? so that when the user's libevent is upgraded it starts working?
146 2017-03-06 17:58:03	0|wumpus|gmaxwell: once my patch lands in mainline libevent, it can be enabled based on the libevent version
147 2017-03-06 17:58:14	0|wumpus|currently there is no way to detect whether libevent can do it or not
148 2017-03-06 18:00:19	0|wumpus|if it takes a long time to be included in libevent we could add the patch to depends so that the released binaries will have it
149 2017-03-06 18:01:12	0|wumpus|can always decide on that when the 0.15 release gets close and it's still uncertain
150 2017-03-06 18:02:15	0|gmaxwell|has the libevent people given any feedback suggesting if they're likely to accept the change or not?
151 2017-03-06 18:02:24	0|wumpus|nope
152 2017-03-06 18:02:52	0|gmaxwell|might be good to get that, if they wont we might want a different approach. :(
153 2017-03-06 18:03:11	0|gmaxwell|(than potentially carrying a patch with release binaries forever)
154 2017-03-06 18:03:19	0|wumpus|well it's still useful even without the client change
155 2017-03-06 18:03:31	0|wumpus|and there is no other approach to support it in bitcoin-cli anyhow
156 2017-03-06 18:03:34	0|BlueMatt|anyone able to reproduce the travis master fail?
157 2017-03-06 18:03:41	0|BlueMatt|verify-commits is failing again but not locally anymore
158 2017-03-06 18:04:05	0|wumpus|not that I know of, at least, maybe libevent people could suggest something, butI haven't had feedback on it at all
159 2017-03-06 18:04:09	0|gmaxwell|BlueMatt: verify commits is failing for me on the 0.14 branch, I haven't checked master today.
160 2017-03-06 18:04:16	0|wumpus|gmaxwell: wow that's depressing :/
161 2017-03-06 18:04:46	0|wumpus|they've had, what, two days including a sunday to respond, I wouldn't worry about "carrying a patch forever" yet
162 2017-03-06 18:05:06	0|BlueMatt|oh wut?
163 2017-03-06 18:06:38	0|wumpus|on 0.14 it should pass now
164 2017-03-06 18:06:51	0|BlueMatt|well it should pass on master too
165 2017-03-06 18:06:58	0|BlueMatt|did it succeed on 0.14 on travis?
166 2017-03-06 18:07:14	0|wumpus|yes I think so
167 2017-03-06 18:07:19	0|BlueMatt|yes, it did
168 2017-03-06 18:59:14	0|BlueMatt|hmmm, MarcoFalke i cant reproduce your Tree-SHA512 from https://github.com/bitcoin/bitcoin/commit/f7ec7cfd38b543ba81ac7bed5b77f9a19739460b
169 2017-03-06 18:59:28	0|BlueMatt|i didnt try the python code, but the bash-based snippet doesnt work here
170 2017-03-06 19:03:52	0|wumpus|verify-commits.sh passes on master here, locally
171 2017-03-06 19:05:58	0|BlueMatt|found the issue...needs gpg 2.1 to pass --weak-digest
172 2017-03-06 19:06:07	0|BlueMatt|also the above tree-sha512 issue
173 2017-03-06 19:07:43	0|wumpus|can't reproduce that one either,tree_sha512sum() gives dc5cf61d226a16a77f32fab8abc18fd469dfe8bda59a2cb8f9eca3a60e474f180dec4fa59bb3fee86efa0e123ba8c198c7efe76bfdaa2518f4169ab0849f6694
174 2017-03-06 19:07:57	0|BlueMatt|yea, that looks like what i (and verify-commits) gets
175 2017-03-06 19:07:59	0|wumpus|it's not causing verify-commits.sh to fail here though
176 2017-03-06 19:08:10	0|BlueMatt|you have to run verify-commits with --tree-checks
177 2017-03-06 19:08:14	0|BlueMatt|(because its slowwwww)
178 2017-03-06 19:08:34	0|BlueMatt|next pr will fix all above issues, and also I think i should do --tree-checks but only for top commit
179 2017-03-06 19:08:39	0|BlueMatt|so that travis will catch this
180 2017-03-06 19:08:50	0|BlueMatt|but not for more because otherwise travis would take hours by the time we get to 0.15
181 2017-03-06 19:09:02	0|wumpus|I agree, we should avoid making the check slower
182 2017-03-06 19:09:23	0|wumpus|checking just the top commit or top two commits thoroughly should be enough, travis runs often enough
183 2017-03-06 19:10:41	0|BlueMatt|yes
184 2017-03-06 19:10:47	0|BlueMatt|it should run every commit, no?
185 2017-03-06 19:10:55	0|BlueMatt|we might see it fail hours later, but at least it runs for each one
186 2017-03-06 19:11:45	0|wumpus|I don't think that's guaranteed. It runs every time it sees a new push
187 2017-03-06 19:12:14	0|BlueMatt|ahh, yes, race
188 2017-03-06 19:12:15	0|wumpus|which usually means it runs for every top-level commit as the merge script pushes them one at a time, but still
189 2017-03-06 19:13:07	0|wumpus|what is it with all those people reporting Ekiga build problems in our issue tracker
190 2017-03-06 19:13:45	0|sipa|ekiga?
191 2017-03-06 19:13:58	0|wumpus|apparently some VOIP program
192 2017-03-06 19:14:17	0|sipa|and what are they reporting? i haven't seen that
193 2017-03-06 19:14:18	0|wumpus|https://github.com/bitcoin/bitcoin/issues/3219#issuecomment-284464031
194 2017-03-06 19:14:34	0|wumpus|I first didn't notice they were talking about a different application
195 2017-03-06 19:15:13	0|gwillen|well, one person happened to get a match on the text of the error message
196 2017-03-06 19:15:24	0|sipa|hah
197 2017-03-06 19:15:29	0|BlueMatt|wtf
198 2017-03-06 19:15:32	0|wumpus|in any case, configure arguments suggestions port very badly between different applications
199 2017-03-06 19:15:33	0|gwillen|then google indexed the word 'ekiga' when they said it, and now everyone's getting linked to this bug when they google 'ekiga' and the error message :-)
200 2017-03-06 19:15:42	0|wumpus|lol that must be it
201 2017-03-06 19:16:33	0|gwillen|indeed, I just checked, that bug is the first hit for the error, and the second hit for 'ekiga' plus the error text.
202 2017-03-06 20:35:35	0|BlueMatt|wumpus: i cant verify your latest sha512 either
203 2017-03-06 20:39:03	0|wumpus|sigh :/
204 2017-03-06 20:40:39	0|wumpus|is it possible that the order is not deterministic?
205 2017-03-06 20:40:42	0|BlueMatt|hmmm, now I'm confused
206 2017-03-06 20:40:45	0|BlueMatt|i doubt it?
207 2017-03-06 20:41:11	0|wumpus|is it hashing any files which are not part of git, but linger around in the directory?
208 2017-03-06 20:41:21	0|BlueMatt|hmm, maybe I'm wrong, maybe your hash is right
209 2017-03-06 20:41:30	0|BlueMatt|cant tell from terminal history anymore :(
210 2017-03-06 20:41:40	0|wumpus|I'll check in a minute, after I push my tree
211 2017-03-06 20:47:23	0|wumpus|for me it matches
212 2017-03-06 20:48:05	0|wumpus|(using the python code, output is e55ce10bf7f2dc91de9797e60ab7767fb51f25255995d62ddf358c52b7aaa23c26fbfb522e1610ff950b86804ddbc38dc0d7708bfab2c4d33ad99a275d8c77db, which matches what is in the merge commit)
213 2017-03-06 20:49:17	0|wumpus|"git ls-tree --full-tree -r --name-only HEAD | LANG=C sort | xargs -n 1 sha512sum | sha512sum" gives the same output
214 2017-03-06 20:49:45	0|BlueMatt|yea, heisenbug
215 2017-03-06 20:49:50	0|BlueMatt|travis is failing on it but i cant find it
216 2017-03-06 21:15:38	0|BlueMatt|wumpus: yes, sipa meant to say LC_ALL when he said LANG (according to the sort man page)
217 2017-03-06 21:25:03	0|sipa|i thought i meant to say C, but maybe i thought wrong about what i meant
218 2017-03-06 21:25:22	0|BlueMatt|sipa: no, you have to set LC_ALL=C, not LANG=C
219 2017-03-06 21:34:06	0|bitcoin-git|[13bitcoin] 15TheBlueMatt opened pull request #9932: Fix verify-commits on travis and always check top commit's tree (06master...062017-03-fix-verify-commits) 02https://github.com/bitcoin/bitcoin/pull/9932
220 2017-03-06 21:35:23	0|sipa|BlueMatt: ahh!
221 2017-03-06 21:38:55	0|wumpus|whoops, yes, stupid locales :/
222 2017-03-06 21:39:23	0|BlueMatt|why travis has a different locale is beside me
223 2017-03-06 21:39:39	0|wumpus|the python version in github-merge.py shouldn't be affected
224 2017-03-06 21:40:56	0|wumpus|no idea what locale it sets, but it helped find this issue
225 2017-03-06 21:41:04	0|wumpus|which I suppose is good
226 2017-03-06 21:41:26	0|BlueMatt|fair
227 2017-03-06 22:05:42	0|bitcoin-git|[13bitcoin] 15jtimon closed pull request #9882: RPC: Introduce -rpcamountdecimals for the RPC to use other units than BTC (06master...060.14.99-rpc-amounts) 02https://github.com/bitcoin/bitcoin/pull/9882
228 2017-03-06 22:06:42	0|jtimon|ping #9279
229 2017-03-06 22:06:45	0|gribble|https://github.com/bitcoin/bitcoin/issues/9279 | Consensus: Move CFeeRate out of libconsensus by jtimon · Pull Request #9279 · bitcoin/bitcoin · GitHub