--- Log for 29.08.121  Server: sodium.libera.chat  Channel: #rockbox
--- Nick: rb-logbot_  Version: Dancer V4.16  Started: 5 days and 23 hours ago

00.41.24 Quit    saratoga (Quit: Connection closed)
00.48.30 ***     Saving seen data "./dancer.seen"
00.59.11 Quit    Moriar (Quit: Leaving.)
02.09.14 Join    dconrad [0] (~dconrad@208.38.228.17)
02.13.40 Quit    dconrad (Ping timeout: 240 seconds)
02.48.31 ***     Saving seen data "./dancer.seen"
04.02.55 Join    lebellium [0] (~lebellium@2a01:cb10:2e:2000:b5cf:6dc:b83d:d90a)
04.09.23 Join    dconrad [0] (~dconrad@208.38.228.17)
04.13.27 Quit    dconrad (Ping timeout: 240 seconds)
04.19.30 Join    ZincAlloy [0] (~Adium@ip5f5abcae.dynamic.kabel-deutschland.de)
04.48.34 ***     Saving seen data "./dancer.seen"
06.09.49 Join    dconrad [0] (~dconrad@208.38.228.17)
06.14.13 Quit    dconrad (Ping timeout: 252 seconds)
06.48.37 ***     Saving seen data "./dancer.seen"
08.30.08 Join    chris_s [0] (~chris_s@ip-95-223-74-182.hsi16.unitymediagroup.de)
08.34.58 #       <chris_s> Is anyone able to test g3769 on a first-generation iPod Mini? It seems to work fine on a Mini 2G but the it's honestly been blindly copied from the Ipod 4G source, specifically this Line:
08.35.02 #       <chris_s> GPIO_SET_BITWISE(GPIOB_ENABLE, 0x0c); /* B02 and B03 enable *
08.46.17 Quit    chris_s (Quit: Connection closed)
08.48.19 #       <speachy> munkis:  Our general needs are met by a $20-30/mo VPS.
08.48.41 ***     Saving seen data "./dancer.seen"
08.50.04 #       <speachy> that's what I will do if I can't find reasonable accomodations for the existing server.  (where "reasonable" means "sufficient bandwidth to be useable for a price I can afford")
08.52.08 #       <speachy> granted, if we completely re-worked _everything_ we could probably fit within someone else's "free project hosting" framework
09.10.31 #       <yang> paulcarroty: I noticed that the headphones jack goes inside m3k with a problem, it gets stuck a bit, so I have to force it....?
09.14.31 #       <yang> munkis: https://fosshost.org/
09.14.51 #       <yang> munkis: this hoster provides free building machines and hosting for Foss projects
09.15.13 #       <yang> munkis: domain is about 10-15 EUR/year
09.45.14 Join    dconrad [0] (~dconrad@208.38.228.17)
09.45.16 Quit    dconrad (Client Quit)
09.59.13 #       <speachy> chris_s:  I don't think anyone active here has a min1g
10.33.48 #       <paulcarroty> yang: slightly
10.46.52 Join    Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net)
10.48.42 ***     Saving seen data "./dancer.seen"
11.35.43 Quit    tchan (Read error: No route to host)
11.39.05 #       <speachy> yang: the "strictly noncommercial" thing is something I'm not comfortable with.
11.41.20 #       <speachy> I will shoot them an inquiry though. 
11.46.40 Join    tchan [0] (~tchan@c-98-206-141-238.hsd1.il.comcast.net)
11.53.46 #       <paulcarroty> yep, don't be shy to ask
12.17.27 Quit    j-r (Ping timeout: 250 seconds)
12.18.08 Join    j-r [0] (~j-r@p2003000621db2c45404207fffefd0a65.dip0.t-ipconnect.de)
12.24.38 Join    amachronic [0] (~amachroni@user/amachronic)
12.48.43 ***     Saving seen data "./dancer.seen"
12.58.31 #       <speachy> the fastest way to shrink our disk space needs are to keep fewer historical daily builds.  we're currently keeping 30 days, and each day is about 2.2GB.
12.58.38 Quit    Moriar (Quit: Leaving.)
13.17.28 #       <braewoods> speachy: rockbox has always been non-profit more or less. why is non-commercial problematic? too vague?
13.17.38 #       <speachy> yeah
13.18.04 #       <speachy> even soliciting donations or having banner ads has been construed to making something "commercial"
13.18.30 #       <speachy> but should we ever end up with our own hardware then that's clearly "commercial" too
13.18.38 #       <braewoods> lmao.
13.18.51 #       <braewoods> to me commercial was always about whether profit is the main motivator
13.19.10 #       <braewoods> if you only seek to cover your operating costs...
13.19.26 #       <braewoods> that's not what i would consider a serious commercial outfit
13.20.25 #       <braewoods> i can't see how rockbox could be seen as commercial; we don't even SELL anything
13.21.02 #       <braewoods> at most we sometimes have developers trade hardware around
13.21.43 #       <braewoods> speachy: what are the requirements you need filled anyway?
13.22.03 #       <speachy> the main sticking point our hosting needs is RAM usage.  gerrit alone sucks down 2GB just sitting there.  :)
13.23.19 #       <speachy> I have some inquiries into some local colo facilities too
13.23.21 #       <braewoods> interesting. i currently rent a soyoustart dedicated server with Romster. 4x 2TB drives and 32GB of ECC RAM.
13.23.35 #       <braewoods> we got it on special for like $32 a month
13.24.23 #       <braewoods> it also includes 16 IPv4 extra addresses we use for containers
13.24.43 #       <speachy> yeah, that's half their current lowest priced offering now
13.25.00 #       <braewoods> i suggest waiting for a black friday special if you're looking at so you start
13.25.09 #       <braewoods> you get the discount for the duration of the lease
13.25.47 #       <braewoods> you also get KVM video access through their web interface
13.25.53 #       <braewoods> you can get into the BIOS and more
13.26.06 #       <braewoods> i actually had to use network booting to get our custom setup installed
13.26.46 #       <braewoods> the ipv4 addresses are really nice for having public containers
13.27.02 #       <braewoods> you pay a one time fee to activate them but the first 16 don't carry a recurring cost
13.27.06 #       <braewoods> it's included in your service
13.27.38 #       <braewoods> containers could be useful for helping separate rockbox systems.
13.27.48 #       <braewoods> i use ubuntu server on our server
13.27.58 #       <braewoods> i have to reboot every 2-3 weeks for kernel upgrades or so
13.27.58 #       <speachy> separating things is more administrative overhead than consolidating everything
13.28.08 #       <braewoods> i see, fair enough.
13.28.43 #       <braewoods> soyoustart also includes 100GB of FTP "backup" with their service
13.28.50 #       <braewoods> it can be upgraded
13.28.57 #       <speachy> do you get to install whatever you want on 'em or does it have to be one of their pre-canned distributions?
13.29.06 #       <braewoods> you can install whatever you want.
13.29.08 #       <speachy> ok.
13.29.11 #       <braewoods> just like on KimSufi.
13.29.22 #       <braewoods> you have full control over the boot process more or less.
13.29.26 #       <speachy> is the backup rsync, or just "dump crap over FTP"
13.29.32 #       <braewoods> dump crap over FTP
13.29.41 #       <speachy> that's kinda useless.  :)
13.29.43 #       <braewoods> i think it has other service options but those cost extra
13.29.46 #       <braewoods> let me look
13.30.20 #       <braewoods> lmao. the us site defaults to French.
13.30.22 #       <braewoods> Why?
13.30.40 #       <bertrik> pourquoi?
13.31.08 #       <speachy> ho.. was engrish for me
13.31.23 #       <braewoods> speachy: ok. the other options are SMBFS or CIFS.
13.31.28 #       <braewoods> err
13.31.30 #       <braewoods> NFS
13.31.32 #       <braewoods> sorry
13.31.46 #       <braewoods> NFS and CIFS
13.32.02 #       <yang> speachy: check also contabo.com & online.net for dedicated servers
13.32.24 #       <braewoods> speachy: the main issue i've seen with OVH is their ipv6 setup is a joke. :)
13.33.01 #       <braewoods> no Prefix Delegation. You have to proxy NDP to get anything to route.
13.34.12 #       <braewoods> speachy: oh, also to note they waive the setup fee if you buy a large enough initial lease.
13.34.20 #       <speachy> I'm practically swimming in servers here.  Just don't have a place to put anything that has any bandwidth to speak of.
13.34.41 #       <braewoods> if you do colo it'd have to be local.
13.34.46 #       <braewoods> otherwise not practical.
13.34.52 #       <speachy> local-ish yeah
13.35.31 #       <braewoods> isn't colo where you just rent the physical space, power, and data? everything else is your problem i presume.
13.35.37 #       <speachy> I have some inquiries, but there's not a lot of colo options that don't involve "capable of withstanding a direct nuke hit" levels of redundancy.
13.36.28 #       <braewoods> well if nuclear war occurs, we've got bigger problems than data preservation :)
13.36.34 #       <yang> speachy: colo is usually more expenssive than renting a powerfull dedicatred server
13.36.39 #       <speachy> yeah
13.37.18 #       <speachy> the smaller operators often will let you just get a couple of U in a shared rack intead of renting an entire cabinet
13.37.46 #       <speachy> we'll see.
13.40.03 #       <speachy> it's a two-pronged problem -- there's rockbox's needs, and then there's my own needs.  rockbox can easily be handled by a moderate VPS.
13.41.12 #       <speachy> throw my needs into the fray (not a lot of bandwidth, but a lot more disk space) and the price jumps a lot.
13.54.16 #       <braewoods> speachy: how much is a lot?
13.55.14 #       <speachy> a VPS that'll handle rockbox's needs can be had for under $30.  Add in 3TB of disk space, and you rapidly hit WTF territory.
13.55.43 #       <braewoods> sounds like dedicated is a better value if you're doing both on the same server.
13.55.51 #       <speachy> yep
13.56.22 #       <braewoods> SYS, the one i rent has 500 megabits i believe
13.56.59 #       <speachy> but the same reason I need that 3TB (and growing) of disk space also means the more cores, the better.  Pre-COVID I'd routinely max everything out for days on end.
13.57.23 #       <braewoods> the SYS server i got uses a quad core... a ...
13.57.34 #       <braewoods> Intel(R) Xeon(R) CPU E3-1231 v3 @ 3.40GHz
13.57.50 #       <braewoods> plus with all the mitigations may actually perform suboptimally
13.57.54 #       <speachy> rockbox's current server has 16 cores.  pre-upgrade it was 12.    :D
13.58.13 #       <braewoods> sounds like you should do colo first then
13.58.41 #       <speachy> but at home here I have 16c/32t box that doesn't do much except when builds are going (and regression tests for my printer stuff)
13.58.57 #       <speachy> that's one of four nodes in this beast; the other three aren't even plugged in
14.00.03 #       <speachy> so I can move the resource-intenstive stuff off of the server, leaving only general-purpose shell server stuff, and then that 4c Xeon E3 is plenty
14.00.24 #       <speachy> (I have one of those lying around too, fwiw..)
14.00.51 #       <speachy> I'm very bandwidth limited by crappy DSL.
14.05.52 #       <paulcarroty> speachy: check ramnode
14.06.53 #       <braewoods> speachy: how bad is the DSL?
14.07.08 #       <braewoods> cable here is pretty sucky for upstream.
14.07.13 #       <braewoods> though 200 megabits down.
14.07.24 #       <speachy> at one location it's 13/1.25 (and actually offline at the moment)
14.07.53 #       <speachy> where I'm typing it's ~40/5 but flaky in crappy weather (which lately is most of the time)
14.07.59 #       <paulcarroty> well, it's feels like 00's beginning
14.09.16 #       <speachy> paying through the nose in both places.
14.27.26 #       <paulcarroty> for data you can use cloud services, guess it will be a lot cheaper
14.48.46 ***     Saving seen data "./dancer.seen"
14.56.51 #       <braewoods> speachy: well our server has a fair bit of unused space. we use it for builds and a minecraft server currently.
14.57.25 #       <braewoods> looks like we use about 200GB out of a 4TB ZFS setup
14.57.55 #       <braewoods> if you want to sub-lease part of it let me know.
14.58.06 #       <braewoods> it sits idle a lot.
14.58.21 #       <braewoods> we already have a web server on the main server front too
15.02.31 #       <braewoods> speachy: there's an option i've known about awhile
15.02.39 #       <braewoods> you heard of OBS?
15.02.54 #       <braewoods> opensuse build service
15.03.03 #       <braewoods> you could probably leverage it to do rockbox test builds
15.03.21 #       <braewoods> provided you could set it up to work within some packaging wrapper
15.04.25 #       <paulcarroty> please don't
15.04.36 Part    f1refly (WeeChat 3.2)
15.04.41 #       <paulcarroty> OBS is gigantic pain in the ass
15.05.10 #       <braewoods> another option is launchpad
15.05.14 #       <paulcarroty> a lot of stupid linters by default
15.05.18 #       <speachy> not worth the hassle.
15.05.22 #       <braewoods> fair enough
15.05.37 #       <speachy> we have builders, and if we lose a few things will just take longer.
15.05.49 #       <braewoods> i'm familiar for both as i maintain the integration with OBS for ungoogled chromium
15.06.54 #       <paulcarroty> me too and I don't wanna use them both after Github Actions experience
15.09.58 Join    Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net)
15.32.05 Join    cockroach [0] (~blattodea@user/cockroach)
16.15.26 #       * speachy sighs.
16.28.27 #       <amachronic> g#3770 if anyone wants to do a quick sanity check
16.28.29 #       <rb-bluebot> 3Gerrit review #3770 at https://gerrit.rockbox.org/r/c/rockbox/+/3770 : 3bugfix: redraw yes/no screen after a full skin update by Aidan MacDonald
16.29.06 #       <amachronic> my concern being whether adding another event might overflow the limited number of event slots available
16.34.52 Join    dconrad [0] (~dconrad@208.38.228.17)
16.38.31 #       <amachronic> though running out of slots is a panic, so I guess I'll merge and see if anyone starts complaining
16.40.14 #       <amachronic> we're at the limit in terms of total event slots used but I don't think the number of concurrent uses is that high.
16.42.42 #       <rb-bluebot> Build Server message: 3New build round started. Revision c11ed99cb4, 303 builds, 10 clients.
16.48.48 ***     Saving seen data "./dancer.seen"
16.49.01 Join    _amachronic [0] (~amachroni@user/amachronic)
16.51.47 Quit    amachronic (Ping timeout: 240 seconds)
16.55.28 #       <rb-bluebot> Build Server message: 3Build round completed after 766 seconds.
16.55.31 #       <rb-bluebot> Build Server message: 3Revision c11ed99cb4 result: All green
17.06.30 Quit    _amachronic (Quit: _amachronic)
17.17.09 Quit    lebellium (Quit: Leaving)
17.21.42 Quit    dconrad (Remote host closed the connection)
17.32.55 Quit    akaWolf (Ping timeout: 252 seconds)
17.33.43 Join    dconrad [0] (~dconrad@208.38.228.17)
17.38.25 Quit    dconrad (Ping timeout: 252 seconds)
17.51.40 Join    akaWolf [0] (~akaWolf@akawolf.org)
18.46.51 Join    advcomp2019_ [0] (~advcomp20@user/advcomp2019)
18.47.09 Join    mendel_munkis [0] (~mendel_mu@ool-ae2cb218.dyn.optonline.net)
18.48.51 ***     Saving seen data "./dancer.seen"
18.49.40 Quit    advcomp2019 (Ping timeout: 240 seconds)
18.49.40 Quit    munkis (Ping timeout: 240 seconds)
18.51.52 Join    edward_ [0] (~edward@user/edward)
18.52.40 Quit    edward (Ping timeout: 240 seconds)
18.54.41 Nick    mendel_munkis is now known as munkis (~mendel_mu@ool-ae2cb218.dyn.optonline.net)
19.14.24 #       <speachy> woo, my remote server came back up of its own accord.  actually rebooted; power must have gone out or something.  So that builder is back online!
19.15.14 #       <speachy> (I'd decided against going to kick it today, now I'm glad I didn't.  It's been offline for nearly two weeks!)
19.20.05 Join    ZincAlloy1 [0] (~Adium@2a02:8108:943f:d824:b1d6:dd9b:1f1b:9345)
19.20.40 Quit    Ckat (Ping timeout: 240 seconds)
19.21.42 Join    Ckat [0] (~Ckat@xn--z7x.xn--6frz82g)
19.22.10 Quit    ZincAlloy (Ping timeout: 240 seconds)
19.40.52 Quit    ZincAlloy1 (Ping timeout: 245 seconds)
20.48.53 ***     Saving seen data "./dancer.seen"
20.52.04 Join    skipwich [0] (~skipwich@user/skipwich)
21.25.29 Quit    cockroach (Quit: leaving)
21.47.34 Quit    Moriar (Ping timeout: 252 seconds)
22.29.33 #       <_bilgus> will targeting c99 specific stuff mess anything up?
22.30.19 #       <speachy> we already use -std=gnu99 so nope.
22.33.29 #       <_bilgus> good!
22.33.32 #       <_bilgus> :)
22.48.57 ***     Saving seen data "./dancer.seen"
23.27.00 Join    Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net)
23.44.27 Quit    Moriar (Ping timeout: 240 seconds)