Le vendredi 19 octobre 2018 à 12:32:40+0200, Markus Gschwendt a écrit : > On Thu, 2018-10-18 at 13:20 +0200, Pierre-Elliott Bécue wrote: > > Le jeudi 18 octobre 2018 à 12:54:20+0200, Markus Gschwendt a écrit : > > > On Thu, 2018-10-18 at 11:02 +0200, Pierre-Elliott Bécue wrote: > > > > Le mardi 16 octobre 2018 à 14:39:38+0200, Markus Gschwendt a écrit : > > > > > > > > > > > ... > > > > What is a "huge memory footprint" when translated to numbers? > > > > > > I would say for a VM only running the list software with a few not too > > > busy lists and exim light 128MB should sufficient. > > > With a mailman-web-gui _not_ more than 250-500MB. > > > > You misunderstood my question. > > Sorry. You are right. I did not understand it that way. > > > How much RAM, according to your friend, does mailman3 use on his system? > > 700MB only for mailman3 > See the Post of him (mailman-users list, Thursday, 13 September 2018 > 2:05 p.m.) > https://lists.mailman3.org/archives/list/mailman-us...@mailman3.org/thr > ead/5LZNBMEMYPZHW5XE7KMPUOAAECK3KP6D/ > > I had a short chat with him yesterday. He didn't answer the comment > of Mark Sapiro because of lack of time at the moment. But in short: > There is not such a thing like 'idle memory'. It is RSS, not virtual > memory. And mailman3 releases the memory not before exiting. > > Myself i was not able to run exim4-light +mailman3 +web-gui packages on > a VM below ~1.8GB RAM. > > My production mailserver exim4-heavy +mailman2 +mailman-web-gui > +dovecot +spamassassin +clamav is running on 3GB RAM, where most of the > memory is used by clamav+sa. Only ~50MB by mailman2.
I'm quite surprised by what you say. My mailman3 instances takes 50MB RSS RAM (it's running since june the 2nd) list 15305 0.0 0.3 172120 50232 ? Ss juin02 0:01 /usr/bin/python3 /usr/lib/mailman3/bin/master -C /etc/mailman3/mailman.cfg The web part takes 76MB RSS RAM. www-data 5082 0.0 0.4 506556 76776 ? Sl juin02 0:01 /usr/bin/uwsgi --ini /etc/mailman3/uwsgi.ini I have no clue of what could explain such a consumption in your case. I'm really open to find solutions as soon as we identify the issue. Cheers! -- Pierre-Elliott Bécue GPG: 9AE0 4D98 6400 E3B6 7528 F493 0D44 2664 1949 74E2 It's far easier to fight for one's principles than to live up to them.
