On Fri, 19 Jun 2020 at 20:46, matthew sporleder <[email protected]> wrote: > > > > > On Jun 19, 2020, at 3:03 PM, Chavdar Ivanov <[email protected]> wrote: > > > > On Fri, 19 Jun 2020 at 16:27, Steffen Nurpmeso <[email protected]> wrote: > >> > >> Chavdar Ivanov wrote in > >> <CAG0OUxjWkyA4MrKYLok-EGJVLS_8=6lb0rd4ped9lu-kydr...@mail.gmail.com>: > >> |I've been watching the discussion with interest, as I am not > >> |particularly verse in these topics; perhaps I've done something not > >> |correctly, but on my 6 years old laptop (4c8t, 20GB memory, core-i7 > >> |3820-qm) the full ' hg clone https://anonhg.netbsd.org/src/' (on a ZFS > >> |placed on an mSATA device) took some 45-50 minutes; the resulting repo > >> |takes about 5GB. I am cloning xsrc right now and will go through a > >> > >> More of that stuff please. This is for you who make their money > >> with managing/creating internet infrastructure! > >> And for all the bored black, white, yellow and red teenagers, > >> their first world internet connection, consuming also while in > >> their Upper East Side apartment. > > > > I beg your pardon, but I could not quite comprehend this remark. My > > Internet connection is reasonably fast - cable 350mb/s - but the > > laptop is beyond a powerline adapter, so that limits the speed to > > about 130mb/s; there is no WiFi involved though. > > > > Anyway, xrsc was also cloned, in three hours and two minutes the build > > from scratch was completed. > > > > I'll retest on my cvs repo on the same machine - but on a SATA SSD - a > > clean build, just to compare the speed difference between zfs and ufs. > > > >> > >> |full build. > >> > >> --steffen > >> | > >> |Der Kragenbaer, The moon bear, > >> |der holt sich munter he cheerfully and one by one > >> |einen nach dem anderen runter wa.ks himself off > >> |(By Robert Gernhardt) > > > > Chavdar > > > > > > -- > > ---- > > I appreciate these works/don’t work but we would need more details from > everyone to make it useful for troubleshooting and finding patterns.
The (admittedly miniscule) data point I offered was that on an old-ish machine with a decent amount of memory, not particularly fast disk ( but using ZFS ) and with reliable and reasonably fast connection it appears to work in a normal time frame, at least on -current amd64, and not stretch to many hours of waiting. But then, I'm pretty sure this is well known - otherwise nobody would have considered Mercurial. That aside, it seems to me ZFS on NetBSD has come to a rather useful state nowadays - I remember when extracting a tar archive into ZFS regularly crashed... -- ----
