|JP Vossen on 14 Sep 2016 16:21:45 -0700|
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
|Re: [PLUG] Replacement mailing list idea|
1) The mailing list & archives (which I, for one, think is critical) 2) The rest of the PLUG web siteIf we need to get off of MCT's infrastructure, I'd think both/all parts need to move. And many thanks to him for all this time!
So for me personally: -100K for Google or Yahoo or anything like that+1 Rich Kulawiec maillist (and I like super-set LUG idea, but...scope for now...)
+1 for a Linode; as others have pointed out they are local, and responsive and generally great, AND at least a few of us on this list already use them BECAUSE of this list & PLUG/Fosscon, so... This would host "the rest of it," and could host the mailing list if something fell through.
or +0.5 for Digital Ocean (.5 only because of the previous Linode notes, D.O. is in NYC and thus local-ish, and I love their docs. But I have a Linode... :)
Rich Kulawiec's comments about spam problems are well founded, though I run my mail relay on my Linode and am not aware of any problems...but it's also not a list server.
Failing the above, I'd trust Lee or Keith's infrastructure, if we wanted to go that route.
My $0.02... On 09/14/2016 06:41 PM, Christopher Barry wrote: > > I had offered to host it on my rackspace box, but this is way better. > > I think this suggestion by Rich is hands down the best solution > available. We get free hosting in a stable environment, backed up > daily, and nurtured by a caring veteran mail system expert on a rock > solid obsd box - what else could anyone want!! Oh, and it's not google, > yahoo, discuss or some other faceless corporation. > > > On Wed, 14 Sep 2016 12:45:49 -0400
Rich Kulawiec <firstname.lastname@example.org> wrote:I sent a couple of comments off-list, so let me send this one on-list. On Wed, Sep 14, 2016 at 10:50:57AM -0400, Rich Freeman wrote:I'd suggest that this be done using a donated service, or by somebody who has significant experience in such things. I know enough about mailing lists and spam filters and such to know that I wouldn't want to treat this task lightly. It isn't rocket science, but we would want competence and stability.I'd be happy to host it here (well, firemountain.net, actually). I'm semi-local (Maryland) after moving from the Philadelphia area, and have been hosting the NoVaLug (Northern Virginia LUG) mailing list for several years, along with a dozen others that have been around much longer. I'm running the latest Mailman 2.X version (2.1.23) on a stable stack of sendmail/OpenBSD. I've done list migrations before (from majordomo 1, majordomo 2, listerv, and Yahoo), so I know how to massage the archives in order to preserve data. And with 30+ years of mailing list and mail system experience, including some very large operations, I think I can handle this. Everything is configured-controlled and backed up daily in order to minimize the possibility of dataloss events and to allow me to back out changes if they don't turn out well. Downtime is minimal: last outage was 3+ months ago when the datacenter hit a perfect storm of extremely high temperatures, a UPS that did, a generator that did, and a cutover relay that didn't. Before that: over a year. I'm evaluating Mailman 3.X. No timetable yet: I'm still kicking the tires and getting used to it. This migration would actually be pretty simple: - clone the list's configuration file from the current one - update for 2.1.23 including DMARC mitigation (ugh) - set new list to temporary moderation - import the current list's "mbox" and rebuild archives - grab the current list roster, check against anti-spam controls, then invite everyone to the new list  - pick a cutover date, on that date turn off the old list, turn off the temp moderation flag on the new one - import the delta between old list's mbox (now) and when last imported (then), add to archives - at this point, the new list is running and has a complete archive of the old list, plus its own ongoing one  Well worth doing. Copious past experience indicates that this is not only a good way of pruning addresses that have been abandoned (but are still accepting mail) but it's also a good way of spotting any false positives in the anti-spam system and quashing them before the new list goes live. ---rsk
Later, JP -- ------------------------------------------------------------------- JP Vossen, CISSP | http://www.jpsdomain.org/ | http://bashcookbook.com/ ___________________________________________________________________________ Philadelphia Linux Users Group -- http://www.phillylinux.org Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug