-
Notifications
You must be signed in to change notification settings - Fork 29.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Release proposal: v2.3.1 #1996
Release proposal: v2.3.1 #1996
Conversation
Noteable changes besides the npm update should be 671e64a, 3806d87 and 671e64a...626432d. |
LGTM |
Tangentially related - would it be okay to transliterate Russian names? Or is this super sensitive? |
@piscisaureus there was conversation about that in #1927, but I don't think a decision was reached. |
@piscisaureus It would be ok from my side. If you do that, you can use «Nikita Skovoroda» (first and last name correspondingly). The current variant was copied from my git settings and/or GitHub user profile. Previous discussion here: #1927. |
Could we also add this to release? #1982 |
that'll have to go in to .mailmap to make it happen @piscisaureus and @ChALkeR |
@bricss That one is not finished yet and has an alternate version in #1998. It requires further discussion and there is no need to delay v2.3.1 release for that, v2.3.1 fixes more serious bugs. #1982/#1998 atm fixes one module using the API in an incorrect and undocumented way (the other one got patched already, there might be more, but I know of only two for sure). I would recommend fixing your problem by merging dtabuenc/karma-html-reporter#27 instead, if you want a fix asap. |
@bricss, This is not the correct place to discuss which of #1998 and #1982 should be merged. |
@piscisaureus, @rvagg, @brendanashworth If the question on how you should write my name is holding the release, just remove those two commits from the above list and put them in some later release when a consensus is reached. Those two commits are insignificant, they target the documentation only. 3806d87 is strictly speaking a security fix and should not be delayed, |
«Nikita Skovoroda» sounds pretty good. |
Updated the commit list in the OP. @rvagg lmk if you're able to get this out today, otherwise I'll see about doing it later today. |
@Fishrock123 This should not be released as-is with the updated commit list. |
@chrisdickinson thoughts on branching to get a 2.3.1 out? I have no problem with it but perhaps there's something special we need to be considering here? @Fishrock123 @ChALkeR also an option is forcing a commit in there and forcing a rebase of everything that's in after and have their shas change. I also don't have a great problem with that but then again I'm a git cowboy. |
... and sorry for being tardy with this, I said I'd release this yesterday but didn't so this is my fault. |
there's only these 3 that would be changed, so perhaps not a big deal? |
Now also:
Imma do a release and force push before any more commits land. Stop me now if you think this is a terrible idea! CI @ 061342a https://jenkins-iojs.nodesource.com/job/iojs+any-pr+multi/54/ |
I've not been following this closely, but it seems #2011 or #2013 should land for the windows path fix to really be complete. cc @Fishrock123 |
I was going on this:
Is that not your reading @silverwind? |
What is happenning now with 2.3.1 brings up a question: why are current patch releases based on the master branch? |
FYI I've opted to sit on this while we further investigate the zlib bug (wasted a chunk of my afternoon doing so!), should get a release out tomorrow, @Fishrock123 || @chrisdickinson are welcome to get this out if they feel the need when they're around before I show up. |
### Notable changes | ||
|
||
3806d87 is strictly speaking a security fix and should not be delayed, | ||
671e64a fixes a problem that, while being a Windows-only bug, caused three bugreports in io.js to the date. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
bug reports
My computer is currently out of service so I'm afraid there's not much I can do to help. :/ Edit: got it working again for a while.. |
Looks like we are too late on this now, we should not put this out until Monday. ((Weekend)) |
agree on putting it off till the end of the weekend, I'm happy to take on board the suggestion to avoid releasing just before or during a weekend to make it easier for people who work in this area as a day-job and don't want weekend drama |
This is the Notable Changes I'm going with, critique welcome.
@targos could you take a look at these since they are mainly yours? |
#2013 is not merged yet, FYI. |
You're right @ChALkeR, I was mixing up |
#2013 is landed CI @ https://jenkins-iojs.nodesource.com/job/iojs+any-pr+multi/66/ Running smoke tests (again) |
For the zlib bug, it is incorrect. |
@targos thanks! What I actually meant to refer to was "buffering" vs "streaming", not "sync" vs "async". I've been tinkering with this bug and so far I'm unable to trigger it under any streaming situation but I'd love to hear if you have different results on this. |
OK, I'm not going to move forward with this release and going to bed instead. There are crypto errors on Windows 2008 that are beyond the random timeouts we might expect. Also, only a few of them are failing with a duration of 60 seconds (the timeout length), so there's something more sinister going on here. @shigeki could this be to do with the headers work, and/or the CNNIC whitelist cert check? /cc @nodejs/crypto
|
https://jenkins-iojs.nodesource.com/job/iojs+pr+win/53/nodes=win2008r2/ is the failing machine, all other slaves are OK, including the Windows 2012 one. Kicking off another one before I head off: https://jenkins-iojs.nodesource.com/job/iojs+any-pr+multi/67/ |
@rvagg hmm, new CI doesn't have those. |
Let's try again then https://jenkins-iojs.nodesource.com/job/iojs+any-pr+multi/68/ I'm still concerned with those failures but I have zero to go on so if anyone has any clues then please fill us in. |
@rvagg still looks green.. |
smoke testing, I've been tuning my smoke test suite and it's coming along nicely! |
success on most fronts, the outstanding smoke test I still can't make play nicely is npm, details here: npm/npm#8648 pushing ahead with release |
PR-URL: #1996 Notable changes * module: The number of syscalls made during a require() have been significantly reduced again (see #1801 from v2.2.0 for previous work), which should lead to a performance improvement (Pierre Inglebert) #1920. * npm: - Upgrade to v2.11.2 (Rebecca Turner) #1956. - Upgrade to v2.11.3 (Forrest L Norvell) #2018. * zlib: A bug was discovered where the process would abort if the final part of a zlib decompression results in a buffer that would exceed the maximum length of 0x3fffffff bytes (~1GiB). This was likely to only occur during buffered decompression (rather than streaming). This is now fixed and will instead result in a thrown RangeError (Michaël Zasso) #1811.
PR-URL: #1996 Notable changes * module: The number of syscalls made during a require() have been significantly reduced again (see #1801 from v2.2.0 for previous work), which should lead to a performance improvement (Pierre Inglebert) #1920. * npm: - Upgrade to v2.11.2 (Rebecca Turner) #1956. - Upgrade to v2.11.3 (Forrest L Norvell) #2018. * zlib: A bug was discovered where the process would abort if the final part of a zlib decompression results in a buffer that would exceed the maximum length of 0x3fffffff bytes (~1GiB). This was likely to only occur during buffered decompression (rather than streaming). This is now fixed and will instead result in a thrown RangeError (Michaël Zasso) #1811.
done, https://iojs.org/dist/latest/ armv6 taking its time as usual, will promote later |
Armv6 is up. 🍻 |
PR-URL #70 Related: nodejs/node#1996 Signed-off-by: Hans Kristian Flaatten <[email protected]>
As pointed out by @silverwind, 671e64a contains an important fix for Windows users that we should get out ASAP. Let me know if there objections to a 24-hour window for this, I figure that a patch release with a limited list of commits should be straightforward.
e56758a5e0
] - async-wrap: add provider id and object info cb (Trevor Norris) #1896d5637e67c9
] - buffer: fix cyclic dependency with util (Brendan Ashworth) #1988c207e8d223
] - build: fix pkg-config output parsing in configure (Ben Noordhuis) #19868d8a26e8f7
] - build: don't run lint from test-ci (Johan Bergström) #19651ec53c044d
] - build: simplify execution of built binary (Johan Bergström) #19553beb880716
] - crypto: add cert check to CNNIC Whitelist (Shigeki Ohtsu) #18956aab2f3b9a
] - deps: make node-gyp work with io.js (cjihrig) iojs/io.js#9903e12561b55
] - deps: upgrade to npm 2.11.2 (Rebecca Turner) #19561f93b63b11
] - doc: change the info to the same as in gitconfig (Christian Tellnes) #20000cf94e6856
] - doc: mention CI in Collaborator Guide (Rich Trott) #19957a3006efe4
] - doc: add TOC links to Collaborator Guide (Rich Trott) #199430638b150f
] - doc: add TSC meeting notes 2015-06-10 (Bert Belder) #1943c4ec04136b
] - doc: reformat authors section (Johan Bergström) #196696165f9be2
] - doc: minor clarification in the modules API doc. (Сковорода Никита Андреевич) #19835c2707c1b2
] - doc: benchmark/README.md copyedit (Rich Trott) #197074fdf732d0
] - doc: copyedit COLLABORATOR_GUIDE.md (Rich Trott) #19645fe6e83640
] - doc: copyedit GOVERNANCE.md (Rich Trott) #1963428526544c
] - doc: add ChALkeR as collaborator (Сковорода Никита Андреевич) #19275dfe0d5d61
] - doc: remove irrelevant SEMVER-MINOR & MAJOR (Rod Vagg)fb8811d95e
] - lib,test: fix whitespace issues (Roman Reiss) #1971a71ee93afe
] - module: reduce syscalls during require search (Pierre Inglebert) #1920671e64ac73
] - module: allow long paths for require on Windows (Michaël Zasso)061342a500
] - net: Defer reading until listeners could be added (James Hartig) #14960abcf44d6b
] - test: add Buffer slice UTF-8 test (Rich Trott) #198988c1831ff4
] - test: tmpdir creation failures should fail tests (Rich Trott) #197652a822d944
] - test: fix test-cluster-worker-disconnect (Santiago Gimeno) #19197c79490bfb
] - test: only refresh tmpDir for tests that need it (Rich Trott) #195488d7904c0b
] - test: remove test repetition (Rich Trott) #187491dfb5e094
] - tools: make test-npm work without global npm (Jeremiah Senkpiel) #19263777f41562
] - tools: enable whitespace related rules in eslint (Roman Reiss) #1971626432d843
] - util: dont repeat isBuffer (Brendan Ashworth) #19881d79f572f1
] - util: move deprecate() to internal module (Brendan Ashworth) #19884b4b1760b5
] - v8: cherry-pick uclibc build patch from upstream (Ben Noordhuis) #19745d0cee46bb
] - vm: remove unnecessary HandleScopes (Ben Noordhuis) #2001953b3e75e8
] - win,node-gyp: enable delay-load hook by default (Bert Belder) iojs/io.js#14333806d875d3
] - zlib: prevent uncaught exception in zlibBuffer (Michaël Zasso) #1811