BALUG-Talk

balug-talk@lists.balug.org

1 participants
154 discussions

Dynamic DNS updates and letsencrypt.org ... yummy! :-)
by Michael Paoli 07 May '20

07 May '20

Dynamic DNS updates and letsencrypt.org ... yummy! :-) So, not too long back, I had a wee little mini-project that gave me motivation to start using, or at to at least seriously consider starting to use, dynamic DNS updates. I'll skip the full details on that, but key bit was, to be able to give specific individual(s), ability to relatively easily and quickly edit DNS data ... but only very specific DNS data. Notably, generally, that of a specific subdomain (Pi.BerkeleyLUG.com) - and any subdomains thereunder. And without giving them ability to alter any other DNS data in or under the zone (BerkeleyLUG.com). Without dynamic DNS updates, there would generally be two possible ways to approach that. One would be to create delegated subdomain. That then fully hands that off, but then burdens them with additional need to have DNS servers, and needing static Internet Protocol addresses (IPs) for those etc., and part of the earlier motivation was that they didn't have static IPs. So that also would burden them with DNS nameservers and need for static IPs - so that would at best be far from idea. Other approach, would be to leverage existing DNS infrastructure, and allow them only to change data in and under the subdomain (Pi.BerkeleyLUG.com) of existing zone (BerkeleyLUG.com). That would be rather to quite difficult, as allowing them to only make DNS changes in the zone file appropriate for the subdomain (and any thereunder), would be a quite non-trivial configuration task. So, such an approach also wasn't very feasible. Not to mention there are generally hazards with allowing folks to edit DNS master zone file(s) - notably errors there can cause significant problems with DNS (up to and including breakage or other issues with up to the entire zone). Well, enter dynamic DNS. With that, I could set up key that could be used only to alter DNS data for the subdomain (and any subdomains thereunder), but that key couldn't change any other DNS data at all. So that was the general approach I pursued. To test, and before enabling dynamic DNS on production (BerkeleyLUG.com) domain/zone, I created a temporary test delegated subdomain, and tested dynamic DNS on that zone (and restricting to a subdomain thereof, and any subdomains thereunder). I then created one more such temporary test delegated subdomain for some further testing. After I was fully satisfied with all that, I moved on to dropping those two temporary test delegated subdomains, and enabling dynamic DNS updates for the production domain/zone, and setting up key that could only change the specifically configured subdomain (Pi.BerkeleyLUG.com) within (and any subdomains thereunder). Anyway, after having that infrastructure in place, I thought of another potentially very useful use case. letsencrypt.org Certificate Authority (CA) TLS(/"SSL") certificates (certs). To validate those, http can be used - but not for wildcard certs, or DNS and including wildcard certs. Since many of the certs I use and prefer (if not "require" - does make things quite a bit simpler) use wildcards, well, DNS validation is the way to go on that. But that's been a quite semi-manual process. Notably, I'd been doing it where requesting client tells me what DNS data to add, I manually add it (add it to zone file and update serial number, reload, wait for that to propagate to the responding authoritative nameservers) ... and do that for each such verification data that needs be added, manually doing each, getting the cert, then removing those temporarily added no longer needed bits of DNS data. Well, certbot(8) (client for letsencrypt.org) has hooks/capabilities to use dynamic DNS updates and/or call programs in place of (or as part of) the "manual" verification (and cleanup) steps. So, by either by giving it access to suitable key (and/)or by having it call programs to handle creating the validation data, those steps can be automated. For DNS, such program(s) would be being given a [sub]domain, and a specific text string, and would then create the relevant TXT record under that [sub]domain. So, I set up a key with the appropriate access to use dynamic DNS (and had earlier set dynamic DNS up across all the zones where I'm master), and set up some helper programs, to allow certbot(8) to leverage the specific key, to make the DNS additions - and also remove them again after validations have completed. So ... generating keys, CSRs, requesting certs, temporarily inserting DNS validation data, getting certs, and removing that temporarily inserted data. That used to be a semi-manual process. The manual part mostly being adding DNS data, checking its propagation, proceeding through that validation step ... for numerous domains across multiple certs. Typically that would take me, oh, maybe roughly 30 minutes or more, maybe once every about 85 days or so (the certs have a lifetime of only 90 days). But with dynamic DNS, and certbot(8) then using the hooks to programs I wrote ... full automated. What used to typically take about 30 minutes or more about every 85 days or so ... much faster and fully automated. Now down to ... under 4 minutes (3m21.050s)! Anyway, I show that below - the/my command used (I did this one against letsencrypt.org test environment, as I don't need to get new certs at present, and test doesn't have the rate limiting of the production environment, and also saves production from having that additional traffic where it's not needed - otherwise production is essentially identical). $ time myCERTBOT_EMAIL= \ > myCERTBOT_OPTS='--preferred-challenges dns --test-cert --server > https://acme-v02.api.letsencrypt.org/directory --manual-auth-hook > mymanual-auth-hook --manual-cleanup-hook mymanual-cleanup-hook' \ > Getcerts \ > '*.balug.org,balug.org,*.archive.balug.org,*.beta.balug.org,*.ipv4.balug.org,*.ipv6.balug.org,*.new.balug.org,*.secure.balug.org,*.staging.balug.org,*.test.balug.org,*.php.test.balug.org,*.wiki.balug.org' > \ > '*.sf-lug.org,sf-lug.org,*.ipv4.sf-lug.org,*.ipv6.sf-lug.org,*.sflug.org,sflug.org,*.sflug.com,sflug.com,*.sflug.net,sflug.net,sf-lug.net,www.sf-lug.net,sf-lug.com,www.sf-lug.com' > \ > 'mpaoli.net,*.mpaoli.net,*.blackie.mpaoli.net,*.old-debian.mpaoli.net,digita…' > \ > '*.balug.org,balug.org,*.lists.balug.org,berkeleylug.com,*.berkeleylug.com' \ > 'berkeleylug.com,*.berkeleylug.com,berkeleylug.org,*.berkeleylug.org' \ > '*.pi.berkeleylug.com,pi.berkeleylug.com'; echo "$?" ... real 3m21.050s user 0m23.092s sys 0m4.706s $ (for f in *_cert.pem; do openssl x509 -noout -text -in "$f" 2>&1 > done) | > sed -ne 's/^ *$Not After :.*$$/\1/p > /DNS:/{s/^ *//;s/, */,/g;s/DNS://g;p}' Not After : Jun 25 16:57:17 2020 GMT *.archive.balug.org,*.balug.org,*.beta.balug.org,*.ipv4.balug.org,*.ipv6.balug.org,*.new.balug.org,*.php.test.balug.org,*.secure.balug.org,*.staging.balug.org,*.test.balug.org,*.wiki.balug.org,balug.org Not After : Jun 25 16:57:24 2020 GMT *.ipv4.sf-lug.org,*.ipv6.sf-lug.org,*.sf-lug.org,*.sflug.com,*.sflug.net,*.sflug.org,sf-lug.com,sf-lug.net,sf-lug.org,sflug.com,sflug.net,sflug.org,www.sf-lug.com,www.sf-lug.net Not After : Jun 25 16:57:28 2020 GMT *.blackie.mpaoli.net,*.digitalwitness.org,*.mpaoli.net,*.old-debian.mpaoli.net,digitalwitness.org,mpaoli.net Not After : Jun 25 16:57:33 2020 GMT *.balug.org,*.berkeleylug.com,*.lists.balug.org,balug.org,berkeleylug.com Not After : Jun 25 16:57:38 2020 GMT *.berkeleylug.com,*.berkeleylug.org,berkeleylug.com,berkeleylug.org Not After : Jun 25 16:57:43 2020 GMT *.pi.berkeleylug.com,pi.berkeleylug.com $ And at the end there, for the certs obtained I show when they expire and the domains/names in their SAN data. Note also in the above, the leading "> " is not literally entered, but PS2 (essentially the shell prompting us that it needs more input to complete the command). So, 6 certs, each with Subject Alternative Name (SAN) having multiple names/domains, and each with at least one, if not more, having wildcard ... all done automatically in 3m21.050s elapsed time. Oh, and another advantage with dynamic DNS updates. Serial numbers handled automatically - one less place for us mere mortal humans to screw up DNS. Also, on the helper programs ... I only wanted the invoking ID to be able to make very certain specific changes to DNS records - the validation records are of a very specific type and format, etc. I could restrict the key a fair bit on that, but not fully to only and exactly just the needed. So, I handle that via the helper programs and a little bit of sudo. The ID doing the request never actually has access to the key itself. Instead, via sudo, the key is used on its behalf - and in program that further checks and restricts, such that only and exactly needed validation data is added (and later removed) - nothing else is allowed. Very feasible that way. Also, the helper programs that add the DNS data - also checks that it's propagated to the responding authoritative nameservers, before continuing (so it should then pass letsencrypt.org's validation tests). But back to, e.g. Pi.BerkeleyLUG.com. Were that to instead be done the "old school" way of editing master zone file, would be quite challenging to set something up that could only update DNS data for that subdomain and subdomains thereunder ... but very easy to limit to only and exactly that on dynamic DNS. So, dynamic DNS updates ... already able to do multiple useful things for me (and other(s)). :-)

1 3

Re: [BALUG-Talk] BALUG list(s) unsubscribe (egad, us listadmins get mail)
by Michael Paoli 02 May '20

02 May '20

"Joseph Perla" <joseph(a)jperla.com> Ah, what listadmins get to deal with :-/ ... (if anyone has constructive criticism on how this process/instructions can be improved, feel free to chime in). I'm sorry, "Joseph Perla" <joseph(a)jperla.com>, but now you're getting annoying. BALUG has over 800 list subscribers, but alas, you're now the less than 1 in 800 that can't figure out how to get the email address you have subscribed, unsubscribed. The instructions from the list and web pages are pretty straight-forward. I also specifically and personally emailed you and the email address subscribed that's apparently getting that subscribed email to you, relevant instructions. Yet it appears you still fail to follow those instructions. It ain't rocket science. I don't know how much more clear it can be. There's a subscribed, opted in (probably double opted in) email address that's receiving email from the list. Great. You say it's some catch-all you can't sent from - whatever, not my problem, that doesn't prevent you (or whomever controls that email) from unsubscribing. By default, you even get monthly password reminder - you shouldn't share that password with anyone (not even me) ... whatever, you also included it in your email. So, okay, go to the web interface - you know what that is, you know what Internet and browser and web are, right? Did you read my email? Did you comprehend it? You put the subscribed email in the box to unsubscribe ... don't have the password, you can request password reminder, and it'll email it to you. So, subscribed email address, password, web form, clicky-clicky, boom, unsubscribed. With that, it's immediate, no email confirmation or the like needed (I think it kicks out one final email that the email address has been unsubscribed) ... and yeah, I just retested all that, works 100% perfectly fine lickety split. So, ain't too damn hard, just friggin' do it. If you can't handle that, give the email address and password to someone else who might volunteer do do that for you - but I'm not going to do that for you. Just do it - get it done - boom, unsubscribed, not that hard. If you can't figure out the instructions, maybe search on the web on how to unsubscribe from a mailman list - probably a few zillion examples/tutorials out there too - it ain't hard. And if that "catch-all" email doesn't want to be unsubscribed, too bad for them - they're apparently forwarding you the list email ... that also allows you to control the subscription of that email ... that's just the way it is. I have absolutely zero control of what email sent to balug.org(a)jperla.com gets forwarded to "Joseph Perla" <joseph(a)jperla.com> Not my problem, if you have issue with that, I suggest you take it up with balug.org(a)jperla.com. Sorry if this sounds frustrated, but well, it is. I get too many of these kinds of darn emails (not frequently, but they do occasionally keep coming up ... relentless occasional dribble/trickle). If you've got useful criticism on how the instructions and such can be improved, sure, send it. If you can provide any diagnostics of where things are failing where they're in fact nor functioning as they ought, great, please provide it - but you've not done any of that. You're basically just "please unsubscribe me". No, it's self-serve - you (or balug.org(a)jperla.com) self-subscribed, and have all the needed access to as, or even more easily, likewise self-unsubscribe the balug.org(a)jperla.com address yourself. So, please do so. Hmmm, hey, I guess this isn't the best time to also put out a pitch, Hey, who'd also like to also volunteer to be a list admin and deal with emails from folks that can't read and follow instructions, on a recurring basis? > From: "Joseph Perla" <joseph(a)jperla.com> > Subject: Re: BALUG list(s) unsubscribe > Date: Sat, 02 May 2020 01:18:38 +0000 > It?s a catch all address and I can?t send email from it. Please unsubscribe > > On Fri, May 1 2020 at 6:16 PM, Michael Paoli < > Michael.Paoli(a)cal.berkeley.edu > wrote: > >> >> >> >> balug.org(a)jperla.com, [Cc: "Joseph Perla" <joseph(a)jperla.com>] >> >> >> >> (Also, "Joseph Perla" <joseph(a)jperla.com>, if your issue is with >> balug.org(a)jperla.com sending/forwarding email to you that you don't want, >> I suggest you take that up with balug.org(a)jperla.com, as balug.org has no >> control over what balug.org(a)jperla.com does with their email). >> >> >> >> Subscribe/unsubscribe is completely self-service on BALUG lists. >> >> >> >> So, you should be able to fully take care of that yourself. >> >> >> >> See the reference information (links, etc.) provided further below. >> >> >> >> Note also: >> o every list mailing contains URL(s) towards end of message body >> to manage subscription(s) - including unsubscribing o in addition to the >> above, every list posting sent also contains >> unsubscribe information within the message headers (notably the >> "List-Unsubscribe: " header contains information on how to >> unsubscribe both via email and via URL) >> >> >> >> If that doesn't work for you, please provide: o The error diagnostic(s) >> (if any) you're getting when you attempt to >> unsubscribe. >> o What steps you're following that result in the above (never tell me/us >> nor anyone your list password, but otherwise provide relevant details) o >> FULL EMAIL HEADERS of an example message that you're getting or still >> getting that you're trying to unsubscribe from. >> >> >> >> Hopefully you should be able to self-unsubscribe, but if that fails, >> please let us know, as noted above. >> >> >> >> Thanks for your attention to this and your cooperation, >> >> >> >> The busy BALUG / list administrator folks. >> >> >> >> See also / references / excerpts / hints: >> https://www.balug.org/#Lists >> https://lists.balug.org/cgi-bin/mailman/listinfo/balug-announce >> https://lists.balug.org/cgi-bin/mailman/listinfo/balug-talk >> https://lists.balug.org/cgi-bin/mailman/listinfo/balug-admin >> https://lists.balug.org/cgi-bin/mailman/listinfo/balug-test >> https://lists.balug.org/cgi-bin/mailman/listinfo >> >> >> >> references/excerpts (redacted): >> >> >>> >>> >>> From: "Joseph Perla" <joseph(a)jperla.com> Subject: Re: lists.balug.org >>> mailing list memberships reminder Date: Fri, 01 May 2020 06:14:36 +0000 >>> >>> >>> >>> Please remove balug.org(a)jperla.com I cannot. >>> >>> >> >> >> >> > >> >> >>> >>> >>> On Thu, Apr 30 2020 at 10:00 PM, < mailman-owner(a)lists.balug.org > wrote: >>> >>> >> >> >> >> > >> >> >> >> >> >> >> >> >>> >>>> >>>> >>>> This is a reminder, sent out once a month, about your lists.balug.org >>>> mailing list memberships. It includes your subscription info and how to >>>> use it to change it or unsubscribe from a list. >>>> >>>> >>> >>> >> >> >> >> >> >> >> >> >> >> >> >>> >>>> >>>> >>>> You can visit the URLs to change your membership status or configuration, >>>> including unsubscribing, setting digest-style delivery or disabling >>>> delivery altogether (e.g., for a vacation), and so on. >>>> >>>> >>> >>> >> >> >> >> >> >> >> >> >> >> >> >>> >>>> >>>> >>>> In addition to the URL interfaces, you can also use email to make such >>>> changes. For more info, send a message to the '-request' address of the >>>> list (for example, mailman-request(a)lists.balug.org) containing just the >>>> word 'help' in the message body, and an email message will be sent to you >>>> with instructions. >>>> >>>> >>> >>> >> >> >> >> >> >> >> >> >> >> >> >>> >>>> >>>> >>>> If you have questions, problems, comments, etc, send them to >>>> mailman-owner(a)lists.balug.org. Thanks! >>>> >>>> >>> >>> >> >> >> >> >> >> >> >> >> >> >> >>> >>>> >>>> >>>> Passwords for balug.org(a)jperla.com: >>>> >>>> >>> >>> >> >> >> >> >> >> >> >> >> >> >> >>> >>>> >>>> >>>> List Password // URL >>>> ---- -------- balug-announce(a)lists.balug.org [REDACTED] >>>> https://lists.balug.org/cgi-bin/mailman/options/balug-announce/balug.org%40…

1 0

test DNS that returns SERVFAIL? ... ! :-)
by Michael Paoli 19 Apr '20

19 Apr '20

test DNS that returns SERVFAIL? ... ! :-) For when one may want a target DNS domain to test against that will generally return SERVFAIL ... I didn't super easily find one out there, so ... (at least for now) created one. ... on the master (IPs for the MNAME in SOA are on this host) $ hostname balug-sf-lug-v2.balug.org $ dig +noall +answer +multiline balug.org. SOA | awk '{if(NR==1)print $5;}' ns0.balug.org. $ dig +short ns0.balug.org. A ns0.balug.org. AAAA 96.86.170.229 2001:470:1f05:19e::2 $ ip a s | egrep 'inet.*(96\.86\.170\.229|2001:470:1f05:19e::2)' inet 96.86.170.229/29 brd 96.86.170.231 scope global eth0 inet6 2001:470:1f05:19e::2/64 scope global $ So, add RR: # nsupdate -l << \. update add servfail.balug.org. 300 IN NS servfail.balug.org. send . # Ah yes, I'm quite starting to get used to and like/prefer dynamic DNS update. Significantly more goof-resistant, and most of the time I don't even have to think about the zone serial number. Which reminds me, I do still want to add some version "control" (tracking) ... driven via cron, so I'll at least have periodic snapshots of changes (since no longer using ye olde manual method & manual version control). For more recent changes, and fine-grained history of changes, logs cover that quite well. But for the longer historical record ... wee bit 'o gap presently to fill on that. Automation is generally a good thing. :-) And with no other (explicit - some DNSSEC automagic bits may be added but we'll ignore those presently) RRs for that domain. So, we then typically get, e.g.: $ dig +noall +answer +comments servfail.balug.org. ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 26642 ;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 1 ;; OPT PSEUDOSECTION: ; EDNS: version: 0, flags:; udp: 4096 $ Note that not all flavors of query against servfail.balug.org. will return SERVFAIL: $ dig +noall +norecurse +answer +authority +comments @ns0.balug.org. servfail.balug.org. ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 20976 ;; flags: qr; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 1 ;; OPT PSEUDOSECTION: ; EDNS: version: 0, flags:; udp: 4096 ;; AUTHORITY SECTION: servfail.balug.org. 300 IN NS servfail.balug.org. $ But in general, trying to do a recursive query on the domain for most RR types, will give SERVFAIL. (Useful for testing, ...)

1 1

Re: [BALUG-Talk] REMINDER: BALUG: meeting: TUESDAY 2020-02-18 Rich Morin...
by aaronco36 17 Feb '20

17 Feb '20

I previously wrote: > Quoting <Michael Paoli Michael.Paoli at cal.berkeley.edu> > > 6:30pm Tuesday, February 18th, 2020 2020-02-18 > > Henry's Hunan Restaurant > > 110 Natoma St. (between 2nd & New Montgomery) > > San Francisco, CA 94105-3704 > > 1-415-546-4999 > > http://henryshunan.com/ > > Easy Transit/Parking Access: short walk from > > BART, MUNI, parking > > Trip planning: http://www.511.org/ > > > > Delicious Hunan cuisine and reasonably priced. > > FYI, http://henryshunan.com/ now links to a > blog by a "Henry Shunan" rather than to the > intended Henry's Hunan Restaurant in SF :-| > It's ultimately up to the BALUG organizer(s) > of course, but could/should perhaps the above > Henry's Hunan link be changed to something like > https://www.yelp.com/biz/henrys-hunan-restaurant-san-francisco > (??) Quoting David Fetter <david(a)fetter.org>: > That's unfortunate. Not to worry; I'm reasonably certain that the principal www.balug.org webmaster(s) will get around to making the fix in the not-too-distant future :-) Also, a commentator on another local LUG mailing-list mentioned that Henry's Hunan Restaurant's current big presence is rather on https://www.facebook.com/HenrysHunanNatoma/ I previously wrote: > Also, and not at all to scare you off, while its > extremely unlikely that this will _in any way > whatsoever_ affect Tuesday evening's BALUG dinner > at Henry's Hunan Restaurant should you decide to > attend it anyway ... Quoting David Fetter <david(a)fetter.org>: > This part is flat-out xenophobic scare-mongering. Stop it. Yeah, sorry about the extra and likely unneeded addition to the original posting :-( -A aaronco36(a)sdf.org ------------------------------------------------------ Welcome to another "Crickets"-free BALUG thread! ------------------------------------------------------

5 5

Re: [BALUG-Talk] REMINDER: BALUG: meeting: TUESDAY 2020-02-18 Rich Morin...
by aaronco36 17 Feb '20

17 Feb '20

Quoting <Michael Paoli Michael.Paoli at cal.berkeley.edu> > 6:30pm Tuesday, February 18th, 2020 2020-02-18 > Henry's Hunan Restaurant > 110 Natoma St. (between 2nd & New Montgomery) > San Francisco, CA 94105-3704 > 1-415-546-4999 > http://henryshunan.com/ > Easy Transit/Parking Access: short walk from > BART, MUNI, parking > Trip planning: http://www.511.org/ > > Delicious Hunan cuisine and reasonably priced. FYI, http://henryshunan.com/ now links to a blog by a "Henry Shunan" rather than to the intended Henry's Hunan Restaurant in SF :-| It's ultimately up to the BALUG organizer(s) of course, but could/should perhaps the above Henry's Hunan link be changed to something like https://www.yelp.com/biz/henrys-hunan-restaurant-san-francisco (??) -A -- Robert, I just don't myself think that the particular discussion topic at this week's BALUG meeting is *primarily* intended for those less-sysadminy, relatively new Linux newcomers like yourself. A better Linux event to drop by to in SF, IMHO, is SF-LUG's actual first-Sunday-of-the-month meeting. Also, and not at all to scare you off, while its extremely unlikely that this will _in any way whatsoever_ affect Tuesday evening's BALUG dinner at Henry's Hunan Restaurant should you decide to attend it anyway -- with the restaurant's delicious and reasonably priced Hunan chicken dishes -- there _have_ been reports by Chinese authorities within just the last two weeks or so confirming a highly pathogenic strain of the H5N1 bird flu in Henry's Hunan province. "Hunan is about 400 km south of Hubei province that has the city of Wuhan, which is the epicentre of coronavirus." (https://theprint.in/health/coronavirus-hit-china-now-reports-h5n1-outbreak-… ) FYI, an excellent web-resource for 95percent-plus Linux events is Rick Moen's Bay Area Linux Events site http://linuxmafia.com/bale/ (the one notable not-just-Linux-or-FOSS exception might be Rick's relatively recent listings for Tom's Berkeley Pi Meeting group http://linuxmafia.com/bale/#berkeleypi) Besides SF-LUG's once-a-month meeting, I'd suggest your also looking into some of the outlier locations for their bona fide _Linux_ end-user support and feedback for those relatively new to installing + using Linux, e.g., BerkeleyLUG, the Diablo Valley Linux User Group (DVLUG), the East Bay Linux User Group (EBLUG), and maybe one or two other regional Linux User Groups. -----

3 3

10 types
by Rick Moen 16 Oct '19

16 Oct '19

At yesterday's BALUG meeting, someone hauled out an old chestnut: There are 10 types of people in this world, those who understand binary and those who don't. -- Unknown I mentioned that there were a couple of responses to that, that I had converted to .signature files. The first was by a poster named Ron Fabre on the Linux Users of Victoria (Melbourne) mailing list (whose comment I .sigged): -- Cheers, There are 10 kinds of people in the world, those who Rick Moen know ternary, those who don't, and those who are now rick(a)linuxmafia.com looking for their dictionaries. -- Ron Fabre McQ! (4x80) Being also on that mailing list, I couldn't help doing meta-commentary: -- Cheers, There are 10 types of people in this world, those who know quaternary, Rick Moen those who only recently figured out Ron Fabre's "ternary" .sig, those who're completely confused, and those who hate self-referential jokes. I also like: There are two kinds of people in the world: Those who can extrapolate from incomplete data and There are 10 kinds of people in the world: those who understand hexadecimal and f the rest. and Q. Why do mathematicians confuse Halloween and Christmas? A. Because 31 Oct = 25 Dec.

3 2

Richard M. Stallman has resigned from FSF and from MIT CSAIL
by Rick Moen 20 Sep '19

20 Sep '19

tl;dr: https://www.fsf.org/news/richard-m-stallman-resigns In the wake of the Jeffrey Epstein matter, controvery arose about dealings between him and the (also now-late) MIT AI scientist Marvin Minsky, and about MIT administration actions related to Epstein. Richard M. Stallman is intimately associated with not just the Free Software Foundation (that he created) but with MIT CSAIL (Computer Science and Artificial Intelligence Laboratory) operation. Fairly recently, there was a discussion on one of CSAIL's unadvertised and privately archived mailing lists, csail-related, where Stallman posted comments.[1] Those comments (mostly about matters peripheral to the late Dr. Minsky) became known to MIT alumna, grad student, and mechanical engineer Selam Jie Gano, who is said to have been sent them by a subscriber. Ms. Gano was very incensed by Stallman's views that she asserted were, among other things, 'defending Epstein'. She publishes an essay on Medium entitled 'Remove Richard Stallman. And everyone else horrible in tech': https://medium.com/@selamie/remove-richard-stallman-fec6ec210794 She also provided what is asserted to be the entire thread to Vice.com's regular 'Motherboard' feature, which published a rather inaccurate piece based largely on her allegations, though providing a curated copy of the mailing list thread with all participant names except Stallman's blacked out. Article: https://www.vice.com/en_us/article/9ke3ke/famed-computer-scientist-richard-… Direct link to Vice.com/Motherboard's curated thread copy: https://www.documentcloud.org/documents/6405929-09132019142056-0001.html?em… Quite a number of other outfits have picked up the story in various ways. Stallman responded by objecting (with perfect justification, as far as I can see) that in no way was he defending Epstein; that this is just not at all what he said. The persons now running Software Freedom Conservancy took advantage of this squalor by 'calling for' Richard to retire from the free software organisation & movement he created (unsigned because they have _that_ little moral backbone). https://sfconservancy.org/news/2019/sep/16/rms-does-not-speak-for-us/ There was also pressure for Richard to resign from his position at CSAIL (where ISTR he's been Visiting Professor). He has now done so, citing 'pressure on MIT' -- along with resigning as President and Board of Directors member at Free Software Foundation. IMO, this affair (e.g., things like the unsigned editorial from the jackals at Software Freedom Conservancy) has opportunism written all over it. One of the less-sucky bits of news/analysis (apparently just before things came to a head) was here: https://fudzilla.com/news/49393-stallman-defends-himself-over-epstein-comme… https://lwn.net/ have not yet caught up with events, and can be expected as usual to be guarded to a fault when they do. https://lists.gnu.org/archive/html/libreplanet-discuss/ has some discussion, but likewise has not caught up with events. [1] None of what I'm writing here imply endorsement of Richard's assertions in the mailing list thread, by the way. In fact, I very much do not concur with some, and would have, among other things, politely corrected his factually incorrect understanding of the legal concept of 'assault'. -- Cheers, Views expressed are my own. Rick Moen rick(a)linuxmafia.com McQ! (4x80)

5 12

Re: [BALUG-Talk] 10 types
by aaronco36 20 Sep '19

20 Sep '19

Rick Moen <rick at linuxmafia.com> wrote: There are 10 kinds of people in the world: those who understand hexadecimal and f the rest. Also saw the following variant: There are 10 kinds of people in the world: the 1 who understands hexadecimal and f the rest. aaronco36(a)sdf.org ---

1 0

Re: [BALUG-Talk] [BALUG-Announce] Re: BALUG: meeting: Tu 2019-09-17 & other BALUG News
by aaronco36 17 Sep '19

17 Sep '19

Another trending topic that tonight's BALUG meeting attendees might consider discussing in-person is that of o RMS's controversial resignation from FSF and from MIT CSAIL. See at least these earlier BALUG-Talk posts before this one, today, at ... - https://lists.balug.org/pipermail/balug-talk/2019-September/000174.html - https://lists.balug.org/pipermail/balug-talk/2019-September/000175.html - https://lists.balug.org/pipermail/balug-talk/2019-September/000176.html aaronco36(a)sdf.org ---------- Forwarded message ---------- Date: Mon, 16 Sep 2019 20:31:53 -0700 From: Michael Paoli <Michael.Paoli(a)cal.berkeley.edu> > REMINDER: BALUG MEETING TOMORROW!: > From: "Michael Paoli" <Michael.Paoli(a)cal.berkeley.edu> > To: BALUG-Announce <balug-announce(a)lists.balug.org> > Subject: BALUG: meeting: Tu 2019-09-17 & other BALUG News > ------------------------------ > For our 2019-09-17 (3rd Tuesday) BALUG meeting: > We don't have a specific speaker/presentation lined up for this meeting, > however ... > Bring topics/questions you'd like to discuss - many interesting > discussions/meetings happen that way. > We may well also discuss some topical item(s) at least some of us have > been working on or poking about recently, e.g.: > o WordPress - site migrations, self-hosting, plugins, backup, ... > o DNS, DNSSEC, AAAA & CAA records, letencrypt.org & certs, etc. > o virtualization & live (host) migrations > o automatic web site updates, shell/scripting languages, crontab/at/... > o Lists, forums, & management thereof > o [L]UG management, dos/don'ts, venues, ...

1 0

Re: [BALUG-Talk] [BALUG-Announce] Reminder: TODAY!: BALUG: meeting: Tu 2019-07-16 Debian! ... & other BALUG News
by Rick Moen 16 Jul '19

16 Jul '19

Quoting Michael Paoli (Michael.Paoli(a)cal.berkeley.edu): >o At this meeting - we'll have a someone with 20 years experience using > and doing systems administration of Debian. Hey, I've been administering Debian since 2.1 'slink', as well, so that makes two decades for me, too. (The Official Debian installer was rather dreadful, then, and the replacement d-i installer introduced with 'sarge' (3.1) is miles better. https://wiki.debian.org/DebianInstaller > >o Debian Birthday ("born" 2019-08-16) / Debian @ 25 / Debian Day / > > Debian Appreciation Day I see that 'Debian Day' aka 'Debian Appreciation Day' is the anniversary of when the late Ian Murdock founded the project on 1993-08-16. Personally, I would want to count from the release date of the first public beta, v. 0.90, on 1994-01-26. Over 25 years either way, though. http://www.ibiblio.org/pub/historic-linux/distributions/debian-0.91/ChangeL…

1 0

← Newer
1
2
3
4
5
6
7
8
...
16
Older →

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

BALUG-Talk