- BALUG-Test - lists.balug.org

test - ignore to: balug-test@lists.balug.org - may not yet work at all nor reliably ...
by Michael Paoli 18 Sep '17

18 Sep '17

test - ignore to: balug-test(a)lists.balug.org - may not yet work at all nor reliably - lists.balug.org only partially configured, various TTLs (notably negative caching) not yet covered, etc.

1 0

Oooh, ... archive tests ... :-)
by Michael Paoli 18 Sep '17

18 Sep '17

So ... since some while (week(s)?) back, I changed the archiving of BALUG-Test from (default of) monthly, to yearly, that's all fine and good *except* ... archiving looks a bit funky ... notably changing that setting doesn't retroactively change what's already been archived, so, e.g. on https://temp.balug.org/pipermail/balug-test/ we see something that looks approximatley like: Archive ... August 2017: ... July 2017: ... 2017: ... Which does look kind'a odd (after all, those months are also within the year 2017 - and it also may not be immedately visually apparent as to whether or not the 2017 archive does or doesn't include July 2017 and August 2017. Also, archive generally lists most recent at top, so, having the most recent (year) data at the bottom is also rather odd in its presentation. So ... let's reload those archives (I'm overdue for some testing of that anyway) ... $ hostname && id && ls -ld /var/lib/mailman/archives/private/balug-test.mbox/balug-test.mbox /var/lib/mailman/archives/private/balug-test.mbox/ /var/lib/mailman/archives/private/ && umask balug-sf-lug-v2.balug.org uid=38(list) gid=38(list) groups=38(list) drwxrws--- 12 www-data list 4096 Aug 6 03:11 /var/lib/mailman/archives/private/ drwxrwsr-x 2 root list 4096 Jul 10 01:13 /var/lib/mailman/archives/private/balug-test.mbox/ -rw-rw-r-- 1 list list 63407 Aug 19 23:53 /var/lib/mailman/archives/private/balug-test.mbox/balug-test.mbox 0077 $ //I'd not altered permissions or used the mailman arch command to (re)load an //archive ... that's the "Debian way" they're set up - at least on //old-stable. Note also that that differs somewhat from what the //mailman check_perms says they ought be ... but regardless it works, //and I'm guestimating for Debian is (moderately?) preferable settings //on those. $ (umask 0114 && /usr/lib/mailman/bin/arch --wipe balug-test) ... And ... https://temp.balug.org/pipermail/balug-test/ that looks *much* better, ... but ... looks can be deceiving, web page looks fine, but web accessible archive bits thereof ... nope - looks like it only has what was already in the "year" archive, but lacks the earlier messages. Ah, ... silly browser (and/or what mailman may have earlier told it regarding caching) ... a reload of the page, and then all looks fine. And ... permissions? Let's see if what resulted looks good/sane in that regard, most notably expecting 'em to match similar archives - or at least be highly compatible (most notably relevant access via the group list). $ (cd /var/lib/mailman/archives/ && find * -follow -type f ! -mmin +60 -exec ls -ld \{\} \;) -rw-rw-r-- 1 list list 1143 Sep 18 01:07 private/balug-test/index.html -rw-rw-r-- 1 list list 31152 Sep 18 01:07 private/balug-test/2017.txt -rw-rw-r-- 1 list list 5036 Sep 18 01:07 private/balug-test/2017/000003.html -rw-rw-r-- 1 list list 6457 Sep 18 01:07 private/balug-test/2017/000005.html -rw-rw-r-- 1 list list 2185 Sep 18 01:07 private/balug-test/2017/000000.html -rw-rw-r-- 1 list list 3804 Sep 18 01:07 private/balug-test/2017/000007.html lrwxrwxrwx 1 list list 11 Sep 18 01:07 private/balug-test/2017/index.html -> thread.html -rw-rw-r-- 1 list list 3449 Sep 18 01:07 private/balug-test/2017/date.html -rw-rw-r-- 1 list list 7136 Sep 18 01:07 private/balug-test/2017/000001.html -rw-rw-r-- 1 list list 2522 Sep 18 01:07 private/balug-test/2017/000008.html -rw-rw-r-- 1 list list 2740 Sep 18 01:07 private/balug-test/2017/000010.html -rw-rw-r-- 1 list list 8441 Sep 18 01:07 private/balug-test/2017/000002.html -rw-rw-r-- 1 list list 8244 Sep 18 01:07 private/balug-test/2017/000004.html -rw-rw-r-- 1 list list 2553 Sep 18 01:07 private/balug-test/2017/000009.html -rw-rw-r-- 1 list list 7865 Sep 18 01:07 private/balug-test/2017/000006.html -rw-rw-r-- 1 list list 3443 Sep 18 01:07 private/balug-test/2017/subject.html -rw-rw-r-- 1 list list 2416 Sep 18 01:07 private/balug-test/2017/000011.html -rw-rw-r-- 1 list list 3390 Sep 18 01:07 private/balug-test/2017/000012.html -rw-rw-r-- 1 list list 3994 Sep 18 01:07 private/balug-test/2017/thread.html -rw-rw-r-- 1 list list 3075 Sep 18 01:07 private/balug-test/2017/000013.html -rw-rw-r-- 1 list list 3445 Sep 18 01:07 private/balug-test/2017/author.html -rw-rw---- 1 list list 1942 Sep 18 01:07 private/balug-test/database/2017-thread -rw-rw---- 1 list list 19497 Sep 18 01:07 private/balug-test/database/2017-article -rw-rw---- 1 list list 2455 Sep 18 01:07 private/balug-test/database/2017-subject -rw-rw---- 1 list list 1756 Sep 18 01:07 private/balug-test/database/2017-date -rw-rw---- 1 list list 2014 Sep 18 01:07 private/balug-test/database/2017-author -rw-rw---- 1 list list 847 Sep 18 01:07 private/balug-test/pipermail.pck -rw-rw-r-- 1 list list 1143 Sep 18 01:07 public/balug-test/index.html -rw-rw-r-- 1 list list 31152 Sep 18 01:07 public/balug-test/2017.txt -rw-rw-r-- 1 list list 5036 Sep 18 01:07 public/balug-test/2017/000003.html -rw-rw-r-- 1 list list 6457 Sep 18 01:07 public/balug-test/2017/000005.html -rw-rw-r-- 1 list list 2185 Sep 18 01:07 public/balug-test/2017/000000.html -rw-rw-r-- 1 list list 3804 Sep 18 01:07 public/balug-test/2017/000007.html lrwxrwxrwx 1 list list 11 Sep 18 01:07 public/balug-test/2017/index.html -> thread.html -rw-rw-r-- 1 list list 3449 Sep 18 01:07 public/balug-test/2017/date.html -rw-rw-r-- 1 list list 7136 Sep 18 01:07 public/balug-test/2017/000001.html -rw-rw-r-- 1 list list 2522 Sep 18 01:07 public/balug-test/2017/000008.html -rw-rw-r-- 1 list list 2740 Sep 18 01:07 public/balug-test/2017/000010.html -rw-rw-r-- 1 list list 8441 Sep 18 01:07 public/balug-test/2017/000002.html -rw-rw-r-- 1 list list 8244 Sep 18 01:07 public/balug-test/2017/000004.html -rw-rw-r-- 1 list list 2553 Sep 18 01:07 public/balug-test/2017/000009.html -rw-rw-r-- 1 list list 7865 Sep 18 01:07 public/balug-test/2017/000006.html -rw-rw-r-- 1 list list 3443 Sep 18 01:07 public/balug-test/2017/subject.html -rw-rw-r-- 1 list list 2416 Sep 18 01:07 public/balug-test/2017/000011.html -rw-rw-r-- 1 list list 3390 Sep 18 01:07 public/balug-test/2017/000012.html -rw-rw-r-- 1 list list 3994 Sep 18 01:07 public/balug-test/2017/thread.html -rw-rw-r-- 1 list list 3075 Sep 18 01:07 public/balug-test/2017/000013.html -rw-rw-r-- 1 list list 3445 Sep 18 01:07 public/balug-test/2017/author.html -rw-rw---- 1 list list 1942 Sep 18 01:07 public/balug-test/database/2017-thread -rw-rw---- 1 list list 19497 Sep 18 01:07 public/balug-test/database/2017-article -rw-rw---- 1 list list 2455 Sep 18 01:07 public/balug-test/database/2017-subject -rw-rw---- 1 list list 1756 Sep 18 01:07 public/balug-test/database/2017-date -rw-rw---- 1 list list 2014 Sep 18 01:07 public/balug-test/database/2017-author -rw-rw---- 1 list list 847 Sep 18 01:07 public/balug-test/pipermail.pck $ //That would, at least at quick glance, appear quite reasonable, and //seems to function okay, etc. //Is it consistent with the other archives (notably in //permissions/ownerships)? $ (cd /var/lib/mailman/archives/ && find */balug-announce -follow -name database -type d -prune -o -type f -exec ls -ld \{\} \;) -rw-rw-r-- 1 list list 1204 Aug 14 01:10 private/balug-announce/index.html -rw-rw-r-- 1 list list 7800 Aug 14 01:10 private/balug-announce/2017-August.txt -rw-rw-r-- 1 list list 3552 Aug 14 03:27 private/balug-announce/2017-August.txt.gz -rw-rw---- 1 list list 856 Aug 14 01:10 private/balug-announce/pipermail.pck -rw-rw-r-- 1 list list 10651 Aug 14 01:10 private/balug-announce/2017-August/000000.html lrwxrwxrwx 1 list list 11 Aug 14 01:10 private/balug-announce/2017-August/index.html -> thread.html -rw-rw-r-- 1 list list 1751 Aug 14 01:10 private/balug-announce/2017-August/date.html -rw-rw-r-- 1 list list 1745 Aug 14 01:10 private/balug-announce/2017-August/subject.html -rw-rw-r-- 1 list list 1772 Aug 14 01:10 private/balug-announce/2017-August/thread.html -rw-rw-r-- 1 list list 1747 Aug 14 01:10 private/balug-announce/2017-August/author.html -rw-rw-r-- 1 list list 1204 Aug 14 01:10 public/balug-announce/index.html -rw-rw-r-- 1 list list 7800 Aug 14 01:10 public/balug-announce/2017-August.txt -rw-rw-r-- 1 list list 3552 Aug 14 03:27 public/balug-announce/2017-August.txt.gz -rw-rw---- 1 list list 856 Aug 14 01:10 public/balug-announce/pipermail.pck -rw-rw-r-- 1 list list 10651 Aug 14 01:10 public/balug-announce/2017-August/000000.html lrwxrwxrwx 1 list list 11 Aug 14 01:10 public/balug-announce/2017-August/index.html -> thread.html -rw-rw-r-- 1 list list 1751 Aug 14 01:10 public/balug-announce/2017-August/date.html -rw-rw-r-- 1 list list 1745 Aug 14 01:10 public/balug-announce/2017-August/subject.html -rw-rw-r-- 1 list list 1772 Aug 14 01:10 public/balug-announce/2017-August/thread.html -rw-rw-r-- 1 list list 1747 Aug 14 01:10 public/balug-announce/2017-August/author.html $ //I'd say that all looks quite excellently consistent. Hopefully our //umask value didn't even matter to the arch command, and it simply "did //the right thing" anyway ... but we set a prudent value - in case arch //didn't set it or explicity set permissions. Maybe I'll (further) test //that at some point ... but that's lower on the priority queue. //And ... let's also test delete/add of a message to make sure that //works as we hope and expect ... //and that all tests good, no surprises there. //So ... onto merging into the archives the older raw mbox //bits obtained from DreamHost.com.

1 0

Digest vs. not issue?
by Michael Paoli 20 Aug '17

20 Aug '17

Saw bit of an apparent(?) issue. I'm wondering if anyone is able to similarly replicate issue. Anyway, I'd subscribed, non-digest form. At some point, switched to digest, ... then back again. After switching to non-digest, email continued to come as digests ... and not just like one following queued-up catch-up one. I tried flipping the setting back and forth, in each case, it always showed the option as I set it, yet somehow I couldn't get the emails to resume coming in non-digest form. That also persisted with bounces of Mailman services, etc. Anyway, unsubscribed, and resubscribed ... hopefully that at least gets me out'a digest mode. And presuming that looks okay, I may later retest again to see if digest mode gets effectively "stuck", when it ought not get stuck in such manner. Anyway, curious if anyone else is able to reproduce that apparent issue I saw.

1 0

Re: [BALUG-Test] BALUG-Test archive --> yearly
by Michael Paoli 18 Aug '17

18 Aug '17

> From: "Rick Moen" <rick(a)linuxmafia.com> > Subject: Re: [BALUG-Admin] list settings? ... > Date: Fri, 18 Aug 2017 06:20:27 -0700 >> Should decide before lists.balug.org. moves over. But other than that >> I think either sounds fine for BALUG-Test. What say ye? >> Preference/recommendation on BALUG-Test? I'm totally open on that >> one. > > Yearly, definitely. > > If the unlikely happens and there's a huge 2017 chunk, all you need do > is change to a different chunk size and regenerate. Nobody is going to > have permalinks, so it doesn't matter. And done - setting changed on BALUG-Test to yearly. If there's any archive bit 'o that that doesn't automagically switch over to yearly, can deal with that sometime before lists.balug.org moves over ... no great rush before then.

1 0

test - ignore
by Michael Paoli 14 Aug '17

14 Aug '17

... from yahoo.com email ... DKIM and/or other possible breakage or issues? Let's find out ...

1 0

Just a From test - ignore
by Michael Paoli 12 Aug '17

12 Aug '17

Just a From test - ignore From postmaster(a)example.com Sat Aug 12 08:04:19 2017 From postmaster(a)example.com Sat Aug 12 08:04:19 2017 > From postmaster(a)example.com Sat Aug 12 08:04:19 2017

1 0

test ignore obscure_address: No
by Michael Paoli 31 Jul '17

31 Jul '17

obscure_address: No

1 1

BALUG-Test: full raw mbox archive publicly available
by Michael Paoli 11 Jul '17

11 Jul '17

And yes, the full raw mbox archive for BALUG-Test is publicly available: $ curl -I https://temp.balug.org/pipermail/balug-test.mbox/balug-test.mbox HTTP/1.1 200 OK Date: Tue, 11 Jul 2017 14:29:02 GMT Server: Apache/2.4.10 (Debian) Last-Modified: Tue, 11 Jul 2017 14:23:34 GMT ETag: "a046-5540b71d48512" Accept-Ranges: bytes Content-Length: 41030 Content-Type: application/mbox $ Procedure documented on: https://www.wiki.balug.org/wiki/doku.php?id=balug:mail_and_lists notably: o Y full mbox archive publicly available; procedure: add: PUBLIC_MBOX = Yes to file: /etc/mailman/mm_cfg.py restart mailman for existing lists, toggling archive from public to private and back again seems sufficient to then create the needed link Also making it available via public rsync is on the todo list, but not towards the higher/highest priority bits. "Not worse than" migration off of DreamHost.com is higher on the priorities. Additional improvements generally come later.

1 0

(pre-announce / "soft" open) BALUG-Test list
by Michael Paoli 11 Jul '17

11 Jul '17

[Bcc: BALUG-Test list] So ... BALUG-Test list ... not an official announcement ... yet, but, it's up and running, and mostly functional ... with some noteworthy caveats/issues still to be addressed (and probably some I've not yet discovered). Policy, etc. - haven't really written that up yet, but shall. In the meantime, it goes approximately as follows: Fairly similar to BALUG-Talk - generally need to be subscribed to post, no job postings (no reason to have exceptions on that on the Test list) It *is* a test list, so some things are and generally will remain different: Recommended convention: if you *do not* want a response to your posting to the list, include both of these two strings: test ignore in your Subject: header and they can be upper and/or lower case (testing should be done case insensitive). If you don't include both test and ignore, you may get human and/or automated responses, e.g. received your posting fine at ... and here's full headers as received. No guarantees you'll get such response(s), but if you don't include test and ignore, you may get such response(s). The list may wink in and out of existence, but for the most part will be persistent, have preserved archives, etc. Archives may be reloaded and/or altered (posts added/dropped) - notably for testing - but for the most part the archives will generally contain what was posted to the list. Anyway, that's at least approximation of the policy (will write it up more later, and have it on the list page). Caveats - at present the archives are effectively broken. It's a web server configuration issue - they're otherwise fine and being archived (and it's not filesystem or file permissions or the like). I'll get that fixed (one of the relatively higher priorities). At present, emails with an empty body are being rejected. I still need to fix that (as legitimate mailman commands may have command in Subject header and have empty body). The mail server has a whole lot 'o anti-spam stuff on it. Fairly likely some of that may be too aggressive and/or annoying. I like how it's rejecting tons 'o cr*p spam :-) ... but I want to keep the nuisance factor sufficiently low on legitimate email. It will probably need some adjusting. If you get a "hard" / "permanent" failure (or the mail server automagically blacklist or bans you on what should be legitimate email), shoot me an email note: Michael.Paoli(a)cal.berkeley.edu - and include full headers and the complete bounce message contents. Some of the greylisting and flood protection may be a bit annoying - especially on first messages. E.g. a legitimate email doesn't necessarily get an "instant" response/acceptance - I've tuned some of that down so initial delays are less annoying (e.g. dropped to 2 minutes) while still effectively thwarting spammers. Anyway, much of that may yet be subject to further adjustments, etc., as reasonably appropriate. There are also a lot of block lists that are used ... at least some of those may be a bit too aggressive in their blocking and block some legitimate email. Anyway, let me know where issues are encountered. Also, the site is neither high bandwidth nor high availability. So, sometimes it may not be available ... but for the most part it generally should be up and available. So, where is it, how to get started? List page, here: https://temp.balug.org/cgi-bin/mailman/listinfo/balug-test It, and the other BALUG lists, will temporarily migrate from lists.balug.org to temp.balug.org - and later settle back on lists.balug.org (part of our migration away from DreamHost.com). BALUG-Test list first, then BALUG-Admin, then BALUG-Talk, and last, BALUG-Announce. And how can you tell if your post was successful without being able to see the archives? Looks like with the default: Receive your own posts to the list? Yes One should be able to tell/distinguish by headers, etc., in received posting. Looks like also there's default: Avoid duplicate copies of messages? Yes One might possibly want to change to No to help ensure all copies are received. And ... well, not exactly step-by-step, but tracking fair bit of migration and testing here: https://www.wiki.balug.org/wiki/doku.php?id=balug:mail_and_lists (it's not fully current ... but not that far from current either)

2 5

test - ignore
by Michael Paoli 10 Jul '17

10 Jul '17

test message - ignore

1 0

2024

2023

2022

2021

2020

2019

2018

2017

BALUG-Test