2025-01-02T14:30:05.691Z | <Matt Benjamin> is there no one available to look into shaman, then? |
2025-01-02T15:14:17.100Z | <Adam Kraitman> I am checking if this failure is related to this PR <https://github.com/ceph/ceph-build/pull/2295> |
2025-01-02T17:36:24.436Z | <Zack Cerza> Curious, what would be the connection there? |
2025-01-02T18:00:07.207Z | <Zack Cerza> Okay, so builds failing with an error like `Failed to evaluate name macro:org.jenkinsci.plugins.tokenmacro.MacroEvaluationException: Error processing tokens` in Jenkins should be working again. This appears to be an issue with a PR I made recently, so I've reverted that change. I don't know if there are separate build issues at the moment. |
2025-01-02T18:03:18.401Z | <Æmerson> There's also something up with chacra or something, it keeps wanting to install ceph-boost and not being able to on at least some machines |
2025-01-02T18:16:42.928Z | <Æmerson> <https://jenkins.ceph.com/job/ceph-api/86812/consoleFull#212386344640526d21-3511-427d-909c-dd086c0d1034> |
2025-01-02T18:16:50.013Z | <Æmerson> I don't know if Chacra took the archive down or what. |
2025-01-02T18:18:15.094Z | <Æmerson> I would honestly like to stop having to deal with ceph-libboost since it makes everything more complicated and we don't do that for other packages or other distros. I know the concern is that we don't want to have to add building boost to every make check run. |
2025-01-02T18:18:34.683Z | <Æmerson> I would honestly like to stop having to deal with ceph-libboost since it makes everything more complicated and breakage prone. We don't do that for other packages or other distros. I know the concern is that we don't want to have to add building boost to every make check run. |
2025-01-02T18:48:31.844Z | <Kyrylo Shatskyy> my vpn stopped working, whenever I try to connect I've got this:
```2025-01-02 19:47:02.085482 VERIFY ERROR: depth=1, error=self-signed certificate in certificate chain: CN=openvpnca-sepia, serial=319824696664668756249066534111061608136050284525
2025-01-02 19:47:02.085622 OpenSSL: error:0A000086:SSL routines::certificate verify failed:
2025-01-02 19:47:02.085684 TLS_ERROR: BIO read tls_read_plaintext error``` |
2025-01-02T18:48:44.793Z | <Kyrylo Shatskyy> is it known failure? |
2025-01-02T18:51:23.216Z | <Æmerson> You need to update a new certificate. |
2025-01-02T18:51:34.722Z | <Æmerson> You need to install a new certificate. |
2025-01-02T18:59:38.846Z | <Kyrylo Shatskyy> I missed something, which one? Where can I get it from? |
2025-01-02T19:46:27.803Z | <Æmerson> <https://filedump.ceph.com/sepia-vpn-client.tar.gz> I think. You need the new certificate for your openvpn configuration. |
2025-01-02T20:09:16.852Z | <Gaurav Sitlani> I notice image rendering issues on <http://ceph.io|ceph.io> , any idea if there's any issue ?: https://files.slack.com/files-pri/T1HG3J90S-F0876EX18AG/download/image.png |
2025-01-02T20:29:10.696Z | <Dan Mick> See the channel topic |
2025-01-02T20:34:17.272Z | <Zack Cerza> I saw 404s attempting to load the images on that page |
2025-01-02T20:34:44.575Z | <Zack Cerza> This is the fix: <https://github.com/ceph/ceph-build/pull/2311> |
2025-01-02T21:11:10.745Z | <Dan Mick> yeah, they're not on the server. |
2025-01-02T21:13:56.203Z | <Dan Mick> still in the source tree. like maybe the most recent "deploy main" job went wrong |
2025-01-02T21:15:26.836Z | <Dan Mick> should probably fix this in [ceph.io](http://ceph.io)'s README: Report any website issues you experience by email to **Mike Perez** [miperez@redhat.com](mailto:miperez@redhat.com), _Ceph Community Manager_. |
2025-01-02T21:20:40.578Z | <Dan Mick> I see odd garbage in the log file for the most recent [www.ceph.io](http://www.ceph.io) PR: |
2025-01-02T21:20:52.684Z | <Dan Mick> ```(node:1780724) UnhandledPromiseRejectionWarning: Error: Command failed with EPIPE: /home/jenkins-build/build/workspace/ceph-website-prs/node_modules/mozjpeg/vendor/cjpeg -quality 75
write EPIPE
����JFIF���ExifMM*V^(�if�HH�0221��0100���(�HH�����Z"��
����}!1AQa"q2���#B��R��$3br� ```
|
2025-01-02T21:21:49.611Z | <Dan Mick> filesystem has space. lots of things in assets/ don't seem to have gotten written. The fact that this command is something to do with images is suspicious |
2025-01-02T21:27:03.542Z | <Zack Cerza> what performs the deploys? |
2025-01-02T21:28:21.038Z | <Dan Mick> Jenkins. There are two jobs ceph-website.. |
2025-01-02T21:29:17.083Z | <Dan Mick> Not sure what does the final promotion but the most recent branch, "revert", also is missing those files. Suspecting JS fuckery |
2025-01-02T21:41:02.783Z | <Dan Mick> yeah, ceph-website-prs writes a test website to <branch>.[ceph.io](http://ceph.io); ceph-website deploys main. Both most-recent jobs have that JFIF garbage |
2025-01-02T21:41:35.514Z | <Zack Cerza> that looks like the first bits of a jpeg file |
2025-01-02T21:42:52.392Z | <Dan Mick> yes |
2025-01-02T21:43:17.612Z | <Dan Mick> something's going horribly wrong with that cjpeg tool, obviously, but I don't see any culprits as to why |
2025-01-02T21:43:21.071Z | <Dan Mick> yet |
2025-01-02T21:44:39.998Z | <Dan Mick> after all the jpeg spew, we get this as well: |
2025-01-02T21:44:51.702Z | <Dan Mick> ``` at WriteWrap.onWriteComplete [as oncomplete] (internal/stream_base_commons.js:92:16)
(Use `node --trace-warnings ...` to show where the warning was created)
(node:1781948) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). To terminate the node process on unhandled promise rejection, use the CLI flag `--unhandled-rejections=strict` (see <https://nodejs.org/api/cli.html#cli_unhandled_rejections_mode>). (rejection id: 2)
(node:1781948) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the Node.js process with a non-zero exit code.```
|
2025-01-02T21:45:09.901Z | <Zack Cerza> `npm audit` on that project is pretty scary i gotta say! |
2025-01-02T21:45:22.970Z | <Dan Mick> like whatever it was writing to died while it was writing |
2025-01-02T21:45:29.930Z | <Dan Mick> I can't spell npm |
2025-01-02T21:45:52.620Z | <Zack Cerza> yeah that made me wonder about security a bit, hence the audit |
2025-01-02T21:46:03.987Z | <Zack Cerza> `153 vulnerabilities (4 low, 85 moderate, 55 high, 9 critical)` |
2025-01-02T21:46:33.139Z | <Dan Mick> that fits my definition of scary (and unmaintained) |
2025-01-02T21:49:15.141Z | <Zack Cerza> i'm curious why dependabot wouldn't be filing PRs about that |
2025-01-02T21:59:24.409Z | <Dan Mick> is there some kind of "fix all those errors" thing that can be tried on a branch? I really don't have a clue |
2025-01-02T22:05:08.164Z | <Dan Mick> npm audit fix, with or without --force, runs into problems with an eleventy plugin |
2025-01-02T22:08:22.196Z | <Zack Cerza> yeah, I was left with
```115 vulnerabilities (2 low, 83 moderate, 28 high, 2 critical)```
so, a _bit_ better? but, this project really does need some actual attention |
2025-01-02T22:10:24.193Z | <Dan Mick> same |
2025-01-02T22:12:46.058Z | <Dan Mick> running npm ci and npm run build:development to see if that jpeg error persists after npm audit without --force |
2025-01-02T22:14:19.436Z | <Dan Mick> didn't get the error, dist/ has assets/bitmaps with files in it |
2025-01-02T22:14:34.068Z | <Dan Mick> the rest of the pr build is just rsync etc. |
2025-01-02T22:14:54.666Z | <Dan Mick> maybe PR the changes to package-lock.json that npm audit fix wrote and see?' |
2025-01-02T22:34:09.980Z | <Kyrylo Shatskyy> thank you, works for me |
2025-01-02T22:53:02.680Z | <Zack Cerza> sure <https://github.com/ceph/ceph.io/pull/771> |
2025-01-02T22:56:19.855Z | <Dan Mick> I almost did the same, but then got worried about "lockfileVersion" |
2025-01-02T22:56:42.061Z | <Dan Mick> I guess the jenkins build will tell us if it's an issue <shrug> |
2025-01-02T22:58:26.204Z | <Dan Mick> on the builder:
$ npm --version
6.14.4 |
2025-01-02T22:58:49.807Z | <Dan Mick> is that why the build failed? <shrug> the error message wouldn't deign to hint about it |
2025-01-02T23:03:50.976Z | <Zack Cerza> that is so amazingly old, i can't even guess what year based on the version number |
2025-01-02T23:04:11.928Z | <Zack Cerza> and... i'm just noticing... we run the PR jobs directly on the production host?? |
2025-01-02T23:05:24.015Z | <Zack Cerza> under a user account with passwordless sudo access |
2025-01-02T23:05:36.562Z | <Zack Cerza> nevermind a code injection vulnerability in a dependency... |
2025-01-02T23:10:14.591Z | <Dan Mick> I'm going to quietly subtract myself from the 'we' in that phrase |
2025-01-02T23:30:18.055Z | <Dan Mick> and, about five years, apparently |