Package server broken, at least five hundred intro cs students affected!

gfb

unread,

Oct 5, 2019, 6:22:30 PM10/5/19

to Racket Users

Could we figure out:

1. What's wrong with the package server's indexing of updated packages, so my students can proceed.

2. An official channel to report problems with the package server.

3. If funding specifically earmarked to support monitoring of and feedback from that channel, fixing of issues, and general reliability work, would help, because I would be happy to contribute some of my professional development funds to that.

Thanks,

Gary Baumgartner

Jesse Alama

unread,

Oct 6, 2019, 8:16:00 AM10/6/19

to Racket Users

Hi Gary,

On Sunday, October 6, 2019 at 12:22:30 AM UTC+2, gfb wrote:

3. If funding specifically earmarked to support monitoring of and feedback from that channel, fixing of issues, and general reliability work, would help, because I would be happy to contribute some of my professional development funds to that.

The core Racket guys can correct me if I'm wrong, but I'd wager that the Racket folks would definitely appreciate some $$$, especially to assist with devops issues such as sites being down, indexing failing, etc. One way to get started would be to set up a recurring donation: https://racket-lang.org/sfc.html . I encourage everyone who relies on Racket for his work or business to help out in this way.

Jesse

Tony Garnock-Jones

unread,

Oct 6, 2019, 10:21:53 AM10/6/19

to Racket Users

Hi Gary,

On Saturday, October 5, 2019 at 11:22:30 PM UTC+1, gfb wrote:

1. What's wrong with the package server's indexing of updated packages, so my students can proceed.

Thanks for the report, and sorry for the delay in responding.

I've made a change that ought to resolve the issue that was preventing the indexer from running, and restarted the indexer. It should be back up to date in the next few minutes.

(The issue was that some of the packages had started to take advantage of a feature of the dependencies list in info.rkt that allowed regular expressions in the platform restriction clause. When it came time for the indexer to convert the dependencies list to JSON, it failed because it didn't know how to handle regular expressions.)

2. An official channel to report problems with the package server.

You found it: emailing me and Jay is as good as it gets at the minute, I'm afraid. We should collectively maybe figure out something better for infrastructural issues!

Regards,

Tony

Bogdan Popa

unread,

Oct 6, 2019, 10:26:13 AM10/6/19

to Tony Garnock-Jones, Racket Users

> (The issue was that some of the packages had started to take advantage of a
> feature of the dependencies list in info.rkt that allowed regular
> expressions in the platform restriction clause. When it came time for the
> indexer to convert the dependencies list to JSON, it failed because it
> didn't know how to handle regular expressions.)

I feel like that was me. Sorry!

gfb

unread,

Oct 6, 2019, 10:46:25 AM10/6/19

to Racket Users

Thank you so much, including for the quick turnaround, on a Sunday no less, once you found out!

Tony Garnock-Jones

unread,

Oct 6, 2019, 10:47:04 AM10/6/19

to Racket Users

On Sunday, October 6, 2019 at 3:26:13 PM UTC+1, Bogdan Popa wrote:

I feel like that was me. Sorry!

It was :-) No apologies please! There was nothing wrong with your packages; it was the package indexer at fault here.

One thing I'd like to do is formalise and then check the contracts that describe the information exchanged among the various infrastructure services. At the moment it's all very dynamic-typey, not even contracted. Introducing machine-checkable descriptions of some of the protocols could give us enough leverage to gradually refactor things into better shape.