From peter at stuge.se Sun Dec 2 21:44:48 2018 From: peter at stuge.se (Peter Stuge) Date: Sun, 2 Dec 2018 21:44:48 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181119161630.12261.qmail@stuge.se> References: <20181114222122.GD16956@my.box> <20181116153650.19409.qmail@stuge.se> <20181119161630.12261.qmail@stuge.se> Message-ID: <20181202214448.3626.qmail@stuge.se> Peter Stuge wrote: > The server is unfortunately stuck under lots of other stuff The good news is that the server is online with 32GB RAM and 256 GB SSD. Please send me your SSH key, then I'll add it. The OS I've set up is a stage4 that I built with catalyst (Gentoo tooling) which isn't intended to run the full system but rather to host containers and/or VMs. There's a systemd with machinectld, but pretty much nothing else. The OS lives on an unencrypted 4GB partition, the rest of the SSD is unused and should be encrypted before it gets mounted in the system. The bad news is that eth0 has now failed, seemingly permanently. This isn't critical, all production communication has always run over eth1, but I have to admit that this is a sign of (too) old hardware. :\ //Peter From nhofmeyr at sysmocom.de Mon Dec 3 15:11:29 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Mon, 3 Dec 2018 16:11:29 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181202214448.3626.qmail@stuge.se> References: <20181114222122.GD16956@my.box> <20181116153650.19409.qmail@stuge.se> <20181119161630.12261.qmail@stuge.se> <20181202214448.3626.qmail@stuge.se> Message-ID: <20181203151129.GA10614@my.box> Some bad news came through on IRC, being that we will not get 850 MHz band spectrum. That's disappointing, but I'm going to continue, naively assuming that 1800 will work out in the end. On Sun, Dec 02, 2018 at 09:44:48PM +0000, Peter Stuge wrote: > Please send me your SSH key, then I'll add it. ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQC+Sir6oqMUNIczrmfTbhj6ZpsRA1FiESlP8tD+9gqMEizTmqGkvv3pdTRamOt/KA98c0AKQa9bwrboALHoBW0IqyAr88hb685G1V16A0lHnKnzI+dMGFpdRgMwYTbzaQmb5exfkE9sslpngMxDTBQI6I3ew+xhs1Eupu/HahDyLN6zixilvXSkN55W1COV00Ue1DgVurVGBRxhLQcHRO49lA2rAZcL421zSHAeViYCcNIlSvS5jwFgDhWbqtmc6H08Iexd68NNYZ6CZBpYciUl8RbyXNurH4JXT5TDA9YCp2hJNAKCK+9bK74RfERi2ZUk97uLxep0oNg+u/W/zuP1 neels at dub5 > The OS I've set up is a stage4 that I built with catalyst (Gentoo tooling) Can I get around having to learn $names? i.e. I just get a root login to a "gsm" host running a bare Debian? (I'd go for 'buster' because that's what I tested the current ansible with, and we got a gcc segfault in osmo-iuh on Debian stable) As soon as I have that root login, I can run the ansible from https://code.fe80.eu/congressgsm/eventgsm and install stuff. branch neels/35c3 There supposedly also is a dev box at AfRA, but last time I checked (after it was fixed) I again couldn't login on it, so it might be down again. So far still working on a local lxc. That's looking pretty good, with .service files installed from osmo-*.git, all services starting up and the eventphone token webservice also basically working against osmo-hlr -- just some issue with/whether-to-use Radis to store the results, but that should be resolved quickly. ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From peter at stuge.se Mon Dec 3 18:24:21 2018 From: peter at stuge.se (Peter Stuge) Date: Mon, 3 Dec 2018 18:24:21 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181203151129.GA10614@my.box> References: <20181114222122.GD16956@my.box> <20181116153650.19409.qmail@stuge.se> <20181119161630.12261.qmail@stuge.se> <20181202214448.3626.qmail@stuge.se> <20181203151129.GA10614@my.box> Message-ID: <20181203182422.26878.qmail@stuge.se> Neels Hofmeyr wrote: > Some bad news came through on IRC, being that we will not get 850 MHz band > spectrum. That's disappointing, but I'm going to continue, naively assuming > that 1800 will work out in the end. I hope 1800 happens. We can get 824-831 MHz (or was it 821-834?) but I don't know whether that's of any use for the femtocells and handsets. > neels at dub5 Added to root at 195.160.172.53 which is the hypervisor. > > The OS I've set up is a stage4 that I built with catalyst (Gentoo tooling) > > Can I get around having to learn $names? i.e. I just get a root login to > a "gsm" host running a bare Debian? You'll have to learn some $names to be able to bring that gsm system up. At the moment there is an unused /dev/sda2 waiting to be set up somehow encrypted. The encryption secret shouldn't be stored on that machine so mounting of that encrypted partition can only be done manually. Once the encrypted partition is mounted where machined looks for images then machinectl can start a tarball or other machined image very easily. We could also consider storing the container(s) in RAM, but that's maybe a bit too risky. :) > (I'd go for 'buster' because that's what I tested the current > ansible with, and we got a gcc segfault in osmo-iuh on Debian stable) Is such a "bare" buster container image available somewhere? //Peter From nhofmeyr at sysmocom.de Mon Dec 3 20:16:14 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Mon, 3 Dec 2018 21:16:14 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181203182422.26878.qmail@stuge.se> References: <20181114222122.GD16956@my.box> <20181116153650.19409.qmail@stuge.se> <20181119161630.12261.qmail@stuge.se> <20181202214448.3626.qmail@stuge.se> <20181203151129.GA10614@my.box> <20181203182422.26878.qmail@stuge.se> Message-ID: <20181203201614.GA22358@my.box> On Mon, Dec 03, 2018 at 06:24:21PM +0000, Peter Stuge wrote: > Is such a "bare" buster container image available somewhere? At least lxc has scripts that install a debian by name, which I use to test the eventgsm ansible. Would it make sense to pull a tar of its root fs?? probably not. Otherwise it is possible to install such debian using debootstrap. I recently tried that and succeeded in placing a bootable debian onto an external HD http://hofmeyr.de/install%20debian%20from%20HD/debootstrap-install.txt Otherwise I'm not familiar with ready images. I could make a fresh file system and debootstrap and then pull an img of that partition?? ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From nhofmeyr at sysmocom.de Tue Dec 4 01:56:07 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Tue, 4 Dec 2018 02:56:07 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181203182422.26878.qmail@stuge.se> References: <20181114222122.GD16956@my.box> <20181116153650.19409.qmail@stuge.se> <20181119161630.12261.qmail@stuge.se> <20181202214448.3626.qmail@stuge.se> <20181203151129.GA10614@my.box> <20181203182422.26878.qmail@stuge.se> Message-ID: <20181204015607.GA8269@my.box> On Mon, Dec 03, 2018 at 06:24:21PM +0000, Peter Stuge wrote: > Is such a "bare" buster container image available somewhere? Step 1: what would be the way to crypt it? - I know cryptmount (which I usually use with an image file, but IIRC works with partitions as well) - Then there's the ext4 encryption feature? - ...? Step 2: install OS would something like this do?? https://us.images.linuxcontainers.org/images/debian/buster/amd64/default/20181203_05:25/rootfs.tar.xz "stolen" from lxc :) Though I see that the lxc images apparently are also put up using debootstrap, so if we can get debootstrap to run manually, that would be preferable I guess. I get this though: osmoserv ~ # emerge dev-util/debootstrap !!! /etc/portage/make.profile is not a symlink and will probably prevent most merges. !!! It should point into a profile within /usr/portage/profiles/ !!! (You can safely ignore this message when syncing. It's harmless.) !!! Your current profile is invalid. If you have just changed your profile !!! configuration, you should revert back to the previous configuration. !!! Allowed actions are limited to --help, --info, --search, --sync, and !!! --version. I have infinite uncertainty there. ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From lynxis at fe80.eu Tue Dec 4 02:14:38 2018 From: lynxis at fe80.eu (Alexander Couzens) Date: Tue, 4 Dec 2018 03:14:38 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181202214448.3626.qmail@stuge.se> References: <20181114222122.GD16956@my.box> <20181116153650.19409.qmail@stuge.se> <20181119161630.12261.qmail@stuge.se> <20181202214448.3626.qmail@stuge.se> Message-ID: <20181204031438.7f1e7a73@lazus.yip> > The good news is that the server is online with 32GB RAM and 256 GB > SSD. Please send me your SSH key, then I'll add it. great! here is mine ssh-rsa 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 lynxis at lazus > The OS I've set up is a stage4 that I built with catalyst (Gentoo > tooling) which isn't intended to run the full system but rather to > host containers and/or VMs. There's a systemd with machinectld, but > pretty much nothing else. The OS lives on an unencrypted 4GB > partition, the rest of the SSD is unused and should be encrypted > before it gets mounted in the system. I don't know machinectl. But I'll read shortly into it. With lxc I would do cryptsetup(LUKS) + lvm and use a lvm volume for the container/vm. > The bad news is that eth0 has now failed, seemingly permanently. This > isn't critical, all production communication has always run over eth1, > but I have to admit that this is a sign of (too) old hardware. :\ How many ethernet ports does it have? Best, lynxis -- Alexander Couzens mail: lynxis at fe80.eu jabber: lynxis at fe80.eu mobile: +4915123277221 gpg: 390D CF78 8BF9 AA50 4F8F F1E2 C29E 9DA6 A0DF 8604 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 833 bytes Desc: OpenPGP digital signature URL: From lynxis at fe80.eu Tue Dec 4 02:23:22 2018 From: lynxis at fe80.eu (Alexander Couzens) Date: Tue, 4 Dec 2018 03:23:22 +0100 Subject: [osmocom-event-orga] good news everyone: BTS and bands In-Reply-To: <0C51C82F-5B1A-4812-9B65-47D7A10FBAAD@ccc.de> References: <20181113160428.GA18310@my.box> <0C51C82F-5B1A-4812-9B65-47D7A10FBAAD@ccc.de> Message-ID: <20181204032322.78b48837@lazus.yip> hi, here a short update on the license. Before calling the bnetza it sounded horrible, that we don't even get any license in the us-850 band. But afterwards it sounded much better. We will hear back in one or two days. I've also talked to Telefonica. But it might to short for this year. I just asked for spectrum in general and wrote later a mail. Best, lynxis -- Alexander Couzens mail: lynxis at fe80.eu jabber: lynxis at fe80.eu mobile: +4915123277221 gpg: 390D CF78 8BF9 AA50 4F8F F1E2 C29E 9DA6 A0DF 8604 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 833 bytes Desc: OpenPGP digital signature URL: From peter at stuge.se Tue Dec 4 23:09:09 2018 From: peter at stuge.se (Peter Stuge) Date: Tue, 4 Dec 2018 23:09:09 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181204031438.7f1e7a73@lazus.yip> <20181204015607.GA8269@my.box> <20181203201614.GA22358@my.box> Message-ID: <20181204230909.24517.qmail@stuge.se> Neels Hofmeyr wrote: > > Is such a "bare" buster container image available somewhere? > > At least lxc has scripts that install a debian by name, which I use > to test the eventgsm ansible. > Would it make sense to pull a tar of its root fs?? probably not. That would work fine, but we have more options. > Step 1: what would be the way to crypt it? > > - I know cryptmount (which I usually use with an image file, but IIRC works > with partitions as well) > - Then there's the ext4 encryption feature? > - ...? Alexander Couzens wrote: > With lxc I would do cryptsetup(LUKS) + lvm and use a lvm volume for > the container/vm. I've done the following: --8<-- One-time setup added /dev/sda2 with all unused SSD space cryptsetup luksFormat /dev/sda2 cryptsetup open /dev/sda2 container mkfs.ext4 /dev/mapper/container cat >> /etc/fstab << EOF /dev/mapper/container /var/lib/container auto noauto 0 0 EOF cryptsetup close container -->8-- --8<-- mounting; these are the manual steps after each reboot cryptsetup open /dev/sda2 container mount /var/lib/container -->8-- > Otherwise it is possible to install such debian using debootstrap. > Otherwise I'm not familiar with ready images. > I could make a fresh file system and debootstrap and then pull an > img of that partition?? .. > Step 2: install OS > > would something like this do?? > https://us.images.linuxcontainers.org/images/debian/buster/amd64/default/20181203_05:25/rootfs.tar.xz Yep! I downloaded it to /var/lib/container/lxcbuster.tar.xz and unpacked into /var/lib/container/lxcbuster/ which then shows up as an image with machinectl: # machinectl list-images NAME TYPE RO USAGE CREATED MODIFIED lxcbuster directory no n/a n/a n/a 1 images listed. # I prepared the unpacked dir: # systemd-machine-id-setup --root /var/lib/container/lxcbuster # echo lxcbuster > /var/lib/container/lxcbuster/etc/hostname Now the container can be started: # machinectl start lxcbuster And a shell opened: # machinectl shell lxcbuster > Though I see that the lxc images apparently are also put up using debootstrap, > so if we can get debootstrap to run manually, that would be preferable I guess. > I get this though: > > osmoserv ~ # emerge dev-util/debootstrap > > !!! /etc/portage/make.profile is not a symlink .. > !!! Your current profile is invalid. .. > I have infinite uncertainty there. Yes, I understand that. The osmoserv system is built by the catalyst tool on a build server (comparable result to debootstrap+ansible, but with complete configurability both during compilation and installation) so I haven't put any package database and not set up any profile. I've installed debootstrap, dpkg and debian-archive-keyring now. (I added them to the spec file on the build server, rebuilt, then copied the binary packages and installed them with qmerge on osmoserv.) Try it out in a new directory in /var/lib/container/ Alexander Couzens wrote: > I don't know machinectl. But I'll read shortly into it. I like it. See also systemd-nspawn, which is container parent. > > The bad news is that eth0 has now failed, seemingly permanently. This > > isn't critical, all production communication has always run over eth1, > > but I have to admit that this is a sign of (too) old hardware. :\ > > How many ethernet ports does it have? Two. The only use for eth0 was to ssh into the system when connected directly with a cable and no keyboard+monitor was availble. It's not critical for operations since all our VLANs come in through eth1. I think I can add a NIC to get more ports. I'll take a look. //Peter From lynxis at fe80.eu Wed Dec 5 00:12:43 2018 From: lynxis at fe80.eu (Alexander Couzens) Date: Wed, 5 Dec 2018 01:12:43 +0100 Subject: [osmocom-event-orga] good news everyone: we have a license for 850!!!! In-Reply-To: <20181204032322.78b48837@lazus.yip> References: <20181113160428.GA18310@my.box> <0C51C82F-5B1A-4812-9B65-47D7A10FBAAD@ccc.de> <20181204032322.78b48837@lazus.yip> Message-ID: <20181205011243.55bff335@lazus.yip> we have a 850 mhz license! yay From nhofmeyr at sysmocom.de Wed Dec 5 12:01:37 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Wed, 5 Dec 2018 13:01:37 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181204230909.24517.qmail@stuge.se> References: <20181204031438.7f1e7a73@lazus.yip> <20181204015607.GA8269@my.box> <20181203201614.GA22358@my.box> <20181204230909.24517.qmail@stuge.se> Message-ID: <20181205120137.GD5760@my.box> On Tue, Dec 04, 2018 at 11:09:09PM +0000, Peter Stuge wrote: > # machinectl start lxcbuster Great! I started it and installed openssh-server, and now I do have a direct ssh login to it from home. Only thing is the name 'lxcbuster', should maybe rather be 'gsmcore'. But that doesn't matter that much, I probably want to wipe and re-create it again at some point. Now installing the current version of https://code.fe80.eu/congressgsm/eventgsm which means most of the collected pubkeys are now on it and most of us should be able to login at 195.160.172.52. See /srv/35c3, there are various READMEs. Things to do: - it needs a second IP address because of the GTP stupidity. If we want to operate both 2G and 3G, both SGSN and GGSN need to be publicly reachable *at the same fixed port number* (defined in spec, unchangeable). - the ansible has some stuff in /etc/network/*, I'm not sure whether that even makes sense to have or needs tweaking. ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From nhofmeyr at sysmocom.de Wed Dec 5 12:10:15 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Wed, 5 Dec 2018 13:10:15 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181205120137.GD5760@my.box> References: <20181204031438.7f1e7a73@lazus.yip> <20181204015607.GA8269@my.box> <20181203201614.GA22358@my.box> <20181204230909.24517.qmail@stuge.se> <20181205120137.GD5760@my.box> Message-ID: <20181205121015.GA8459@my.box> On Wed, Dec 05, 2018 at 01:01:37PM +0100, Neels Hofmeyr wrote: > Things to do: - osmo-hnbgw says 20181205130759882 DMAIN NOTICE Listening for Iuh at 195.160.172.52 29169 (hnbgw.c:602) 20181205130759882 DLGLOBAL ERROR unable to create socket: Protocol not supported (socket.c:97) 20181205130759882 DLGLOBAL ERROR no suitable addr found for: 195.160.172.52:29169 (socket.c:378) Cannot open server: Protocol not supported Any reason why binding an SCTP port wouldn't work? (It works on my local lxc, apparently) Maybe the image is using the host's kernel that has no SCTP support? ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From peter at stuge.se Wed Dec 5 13:06:43 2018 From: peter at stuge.se (Peter Stuge) Date: Wed, 5 Dec 2018 13:06:43 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> Message-ID: <20181205130643.6927.qmail@stuge.se> Neels Hofmeyr wrote: > I started it and installed openssh-server, and now I do have a direct ssh > login to it from home. > > Only thing is the name 'lxcbuster', should maybe rather be 'gsmcore'. But > that doesn't matter that much, I probably want to wipe and re-create it > again at some point. I thought it might just be a first example and that debootstrap (installed last night) would be used to set up the real container. Anyway the name is simply the directory name in /var/lib/container so changing it is easy. (But better power off the container first.) > Things to do: > > - it needs a second IP address because of the GTP stupidity. > If we want to operate both 2G and 3G, both SGSN and GGSN need to be publicly > reachable *at the same fixed port number* (defined in spec, unchangeable). I've added 195.160.172.51. > - the ansible has some stuff in /etc/network/*, I'm not sure whether that > even makes sense to have or needs tweaking. The 195.160. IPs are obviously only valid at CCCB. Maybe it's better to set them in a file in /srv/35c3/etc/network/interfaces.d/ instead of /etc/network/interfaces as I did? > - osmo-hnbgw says > > 20181205130759882 DLGLOBAL ERROR unable to create socket: Protocol not supported (socket.c:97) > > Any reason why binding an SCTP port wouldn't work? > Maybe the image is using the host's kernel that has no SCTP support? Yes, sorry, I forgot about that in the kernel config. I've added SCTP (the kernel tree is under /root/ - run make then lilo) and rebooted - it looks good. # systemctl status ? gsmcore State: running Jobs: 0 queued Failed: 0 units //Peter From nhofmeyr at sysmocom.de Thu Dec 6 02:10:21 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Thu, 6 Dec 2018 03:10:21 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181205130643.6927.qmail@stuge.se> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> Message-ID: <20181206021021.GB27230@my.box> On Wed, Dec 05, 2018 at 01:06:43PM +0000, Peter Stuge wrote: > I thought it might just be a first example and that debootstrap > (installed last night) would be used to set up the real container. ah ok. Yes, why not. So far it works pretty good though :) > I've added 195.160.172.51. yay! > > - the ansible has some stuff in /etc/network/*, I'm not sure whether that > > even makes sense to have or needs tweaking. > > The 195.160. IPs are obviously only valid at CCCB. Maybe it's better > to set them in a file in /srv/35c3/etc/network/interfaces.d/ instead > of /etc/network/interfaces as I did? yeah, there's both an /etc/network/interfaces as well as an interfaces.d in the eventgsm.git, but which links get set is defined in the setup-osmocom.yml file. I think I have consciously omitted /etc/network/interfaces so far. > I've added SCTP (the kernel tree is under /root/ - run make then lilo) > and rebooted - it looks good. Confirmed! Excellent stuff. We should have the first tests running on it by this weekend. I still disliked how the MNCC socket is configured and hence wrote some osmo-msc patches, so it can be set in the .cfg and doesn't require the -M cmdline option. They are already merged, yay! rebuilding... done, works. Also going to give Thomas from POC access once SIP works, so he can connect up his BTS and test codecs. ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From peter at stuge.se Thu Dec 6 17:00:26 2018 From: peter at stuge.se (Peter Stuge) Date: Thu, 6 Dec 2018 17:00:26 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181206021021.GB27230@my.box> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181206021021.GB27230@my.box> Message-ID: <20181206170026.13045.qmail@stuge.se> Neels Hofmeyr wrote: > > I thought it might just be a first example and that debootstrap > > (installed last night) would be used to set up the real container. > > ah ok. Yes, why not. So far it works pretty good though :) If it's good enough then we can rename it when convenient. > > I've added 195.160.172.51. > > yay! Do we need to do some routing tricks for everything to work right? > I think I have consciously omitted /etc/network/interfaces so far. I think that's wise, because IPs are a bit more volatile than the network configuration. > We should have the first tests running on it by this weekend. .. > I still disliked how the MNCC socket is configured and hence wrote some > osmo-msc patches, so it can be set in the .cfg and doesn't require the -M > cmdline option. They are already merged, yay! rebuilding... done, works. Great! > Also going to give Thomas from POC access once SIP works, so he can > connect up his BTS and test codecs. Let me know if I can help with SIP. //Peter From nhofmeyr at sysmocom.de Sat Dec 8 16:15:33 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Sat, 8 Dec 2018 17:15:33 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181206170026.13045.qmail@stuge.se> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181206021021.GB27230@my.box> <20181206170026.13045.qmail@stuge.se> Message-ID: <20181208161533.GA16513@my.box> On Thu, Dec 06, 2018 at 05:00:26PM +0000, Peter Stuge wrote: > Let me know if I can help with SIP. I have another problem: Dec 08 17:10:28 gsmcore osmo-stp[63]: 20181208171028346 DLINP DEBUG accept()ed new link from 127.0.0.1 to port 2905 (stream.c:585) Dec 08 17:10:28 gsmcore osmo-stp[63]: 20181208171028347 DLINP ERROR coudldn't activate SCTP events on FD 7 (stream.c:95) osmo-stp somehow can't use SCTP, so we have no connection between BSC and MSC. I have no idea how to fix that... ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From nhofmeyr at sysmocom.de Sat Dec 8 18:20:46 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Sat, 8 Dec 2018 19:20:46 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181205130643.6927.qmail@stuge.se> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> Message-ID: <20181208182046.GB16513@my.box> On Wed, Dec 05, 2018 at 01:06:43PM +0000, Peter Stuge wrote: > > Any reason why binding an SCTP port wouldn't work? > > Maybe the image is using the host's kernel that has no SCTP support? > > Yes, sorry, I forgot about that in the kernel config. > > I've added SCTP (the kernel tree is under /root/ - run make then lilo) > and rebooted - it looks good. I'm getting this error and osmo-stp cannot accept connections: Dec 08 19:06:13 gsmcore osmo-stp[63]: 20181208190613528 DLINP DEBUG accept()ed new link from 127.0.0.1 to port 2905 (stream.c:585) Dec 08 19:06:13 gsmcore osmo-stp[63]: 20181208190613528 DLINP ERROR coudldn't activate SCTP events on FD 7 (stream.c:95) Could it be a missing kernel config? Or maybe we need to install some sctp library on the gentoo host? ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From peter at stuge.se Sun Dec 9 03:54:27 2018 From: peter at stuge.se (Peter Stuge) Date: Sun, 9 Dec 2018 03:54:27 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181208182046.GB16513@my.box> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> Message-ID: <20181209035427.17214.qmail@stuge.se> Hi, Neels Hofmeyr wrote: > I'm getting this error and osmo-stp cannot accept connections: > > Dec 08 19:06:13 gsmcore osmo-stp[63]: 20181208190613528 DLINP DEBUG accept()ed new link from 127.0.0.1 to port 2905 (stream.c:585) > Dec 08 19:06:13 gsmcore osmo-stp[63]: 20181208190613528 DLINP ERROR coudldn't activate SCTP events on FD 7 (stream.c:95) This is due to a struct sctp_event_subscribe definition mismatch between --8<-- gsmcore:/usr/include/linux/sctp.h:582 struct sctp_event_subscribe { __u8 sctp_data_io_event; __u8 sctp_association_event; __u8 sctp_address_event; __u8 sctp_send_failure_event; __u8 sctp_peer_error_event; __u8 sctp_shutdown_event; __u8 sctp_partial_delivery_event; __u8 sctp_adaptation_layer_event; __u8 sctp_authentication_event; __u8 sctp_sender_dry_event; __u8 sctp_stream_reset_event; __u8 sctp_assoc_reset_event; __u8 sctp_stream_change_event; }; -->8-- ( which gets used at /srv/35c3/src/libosmo-netif/src/stream.c:76 ) and the osmoserv kernel struct --8<-- osmoserv:/root/linux*1/include/uapi/linux/sctp.h:484 struct sctp_event_subscribe { __u8 sctp_data_io_event; __u8 sctp_association_event; __u8 sctp_address_event; __u8 sctp_send_failure_event; __u8 sctp_peer_error_event; __u8 sctp_shutdown_event; __u8 sctp_partial_delivery_event; __u8 sctp_adaptation_layer_event; __u8 sctp_authentication_event; __u8 sctp_sender_dry_event; }; -->8-- The osmoserv kernel code looks like this: --8<-- static int sctp_setsockopt_events(struct sock *sk, char __user *optval, unsigned int optlen) { struct sctp_association *asoc; struct sctp_ulpevent *event; if (optlen > sizeof(struct sctp_event_subscribe)) return -EINVAL; ... -->8-- The three additional struct members in the debian header file cause the above struct size check to fail. (Please remember to include the kernel errno in userspace error messages. In this case the issue is clear anyway, but that's not always the case.) The linux-libc-dev debian package which provides the header file is based on debian kernel version 4.18.20-2, while the osmoserv kernel was a 4.9 longterm version. Somewhat reluctantly (real bad block layer bug fixed only the other day) I've updated osmoserv to 4.19.8, that has all struct members in the debian header, so osmo-stp should just work now. //Peter From nhofmeyr at sysmocom.de Mon Dec 10 00:24:03 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Mon, 10 Dec 2018 01:24:03 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181209035427.17214.qmail@stuge.se> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> Message-ID: <20181210002403.GA28498@my.box> On Sun, Dec 09, 2018 at 03:54:27AM +0000, Peter Stuge wrote: > Somewhat reluctantly (real bad block layer bug fixed only the other day) > I've updated osmoserv to 4.19.8, that has all struct members in the > debian header, so osmo-stp should just work now. Took a look now and it works! I can now connect a BTS to our gsmcore and get SIP on the other end, going towards the POC dev machine (but SIP is not working yet). POC says we can sort out connecting to SIP on tuesday. ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From peter at stuge.se Mon Dec 10 03:05:54 2018 From: peter at stuge.se (Peter Stuge) Date: Mon, 10 Dec 2018 03:05:54 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181210002403.GA28498@my.box> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> Message-ID: <20181210030554.8900.qmail@stuge.se> Neels Hofmeyr wrote: > > Somewhat reluctantly (real bad block layer bug fixed only the other day) > > I've updated osmoserv to 4.19.8, that has all struct members in the > > debian header, so osmo-stp should just work now. > > Took a look now and it works! > I can now connect a BTS to our gsmcore and get SIP on the other end, > going towards the POC dev machine (but SIP is not working yet). Good news. Thanks for confirming! > POC says we can sort out connecting to SIP on tuesday. All right. If you need it, sngrep is installed on osmoserv. Soon we'll have to plan how to transport the machine to Leipzig. One opportunity is now on Friday, the 14:th, but I think there will be at least one later opportunity, by piggybacking on VOC stuff going from CCCB. //Peter From nhofmeyr at sysmocom.de Mon Dec 10 11:49:49 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Mon, 10 Dec 2018 12:49:49 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181210030554.8900.qmail@stuge.se> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> <20181210030554.8900.qmail@stuge.se> Message-ID: <20181210114949.GA1991@my.box> On Mon, Dec 10, 2018 at 03:05:54AM +0000, Peter Stuge wrote: > Soon we'll have to plan how to transport the machine to Leipzig. One > opportunity is now on Friday, the 14:th, but I think there will be at > least one later opportunity, by piggybacking on VOC stuff going from > CCCB. I'll be driving with a camper on 22nd, that could be the emergency fallback. Has anyone sorted out transport of the event pool boxes from sysmocom office yet? ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From peter at stuge.se Mon Dec 10 12:17:25 2018 From: peter at stuge.se (Peter Stuge) Date: Mon, 10 Dec 2018 12:17:25 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181210114949.GA1991@my.box> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> <20181210030554.8900.qmail@stuge.se> <20181210114949.GA1991@my.box> Message-ID: <20181210121725.24654.qmail@stuge.se> Neels Hofmeyr wrote: > > Soon we'll have to plan how to transport the machine to Leipzig. One > > opportunity is now on Friday, the 14:th, but I think there will be at > > least one later opportunity, by piggybacking on VOC stuff going from > > CCCB. > > I'll be driving with a camper on 22nd, that could be the emergency fallback. That's a great option! Thanks a lot. > Has anyone sorted out transport of the event pool boxes from sysmocom office > yet? If they are at CCCB by Thursday evening I think they can go in that first truck. Unfortunately I can't help with transport to CCCB, but I can make sure that someone will receive them there. //Peter From nhofmeyr at sysmocom.de Tue Dec 11 00:30:50 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Tue, 11 Dec 2018 01:30:50 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181210121725.24654.qmail@stuge.se> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> <20181210030554.8900.qmail@stuge.se> <20181210114949.GA1991@my.box> <20181210121725.24654.qmail@stuge.se> Message-ID: <20181211003050.GF6861@my.box> On Mon, Dec 10, 2018 at 12:17:25PM +0000, Peter Stuge wrote: > > I'll be driving with a camper on 22nd, that could be the emergency fallback. > > That's a great option! Thanks a lot. > > > > Has anyone sorted out transport of the event pool boxes from sysmocom office > > yet? > > If they are at CCCB by Thursday evening I think they can go in that Ok, lynxis has confirmed that he will take over responsibility for the event pool boxes, also about inventorizing and insurance via ccc event gmbh. He will apparently take the boxes from the office on Saturday or Sunday. The gsmcore machine could probably go along with those boxes, too? Otherwise for all of this, me driving there on 22nd is another possibility. So, Peter, if you can get the gsmcore box to the sysmocom office on/before Saturday (pending lynxis' approval), or to my place (Kreuzberg) on/before 22nd, then that should work out. ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From nhofmeyr at sysmocom.de Tue Dec 11 10:48:28 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Tue, 11 Dec 2018 11:48:28 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181211003050.GF6861@my.box> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> <20181210030554.8900.qmail@stuge.se> <20181210114949.GA1991@my.box> <20181210121725.24654.qmail@stuge.se> <20181211003050.GF6861@my.box> Message-ID: <20181211104828.GH6861@my.box> On Tue, Dec 11, 2018 at 01:30:50AM +0100, Neels Hofmeyr wrote: > So, Peter, if you can get the gsmcore box to Or tell me where to pick it up, but it would be an effort to drive around town in a huge oldtimer. ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From peter at stuge.se Wed Dec 12 14:58:51 2018 From: peter at stuge.se (Peter Stuge) Date: Wed, 12 Dec 2018 14:58:51 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181211003050.GF6861@my.box> References: <20181205121015.GA8459@my.box> <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> <20181210030554.8900.qmail@stuge.se> <20181210114949.GA1991@my.box> <20181210121725.24654.qmail@stuge.se> <20181211003050.GF6861@my.box> Message-ID: <20181212145851.27003.qmail@stuge.se> Neels Hofmeyr wrote: > .. apparently take the boxes from the office on Saturday or Sunday. > The gsmcore machine could probably go along with those boxes, too? We discussed this today - the plan is to ship the server out of CCCB on Friday along with other stuff from there - except for the SSD, which travels with lynxis on Sunday. I've just modified gsmcore:/etc/ssh/sshd_config to disable password authentication, but there was no risk so far since no users have passwords set. I'll shut the machine down on Friday for pickup, but not before, in case there are further configuration changes. //Peter From nhofmeyr at sysmocom.de Wed Dec 12 16:01:04 2018 From: nhofmeyr at sysmocom.de (Neels Hofmeyr) Date: Wed, 12 Dec 2018 17:01:04 +0100 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181212145851.27003.qmail@stuge.se> References: <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> <20181210030554.8900.qmail@stuge.se> <20181210114949.GA1991@my.box> <20181210121725.24654.qmail@stuge.se> <20181211003050.GF6861@my.box> <20181212145851.27003.qmail@stuge.se> Message-ID: <20181212160104.GH1926@my.box> On Wed, Dec 12, 2018 at 02:58:51PM +0000, Peter Stuge wrote: > I've just modified gsmcore:/etc/ssh/sshd_config to disable password > authentication, but there was no risk so far since no users have > passwords set. Absorbed that into the eventgsm.git. Another thing that would be good is to set everything up as non-root, or at least run the services as non-root. But leaving that for later / next year. > I'll shut the machine down on Friday for pickup, but not before, in > case there are further configuration changes. Thomas from POC has just started testing voice last night, using a VPN to connect his lime to our gsmcode. Might come back today, let's see. Last night I even got a friendly woman telling me that "ngxs number is currently not assigned", so AFAICT we're already properly connected to the POC yate. Just need to test voice codecs, test the token interaction, dump all sim card data in the hlr, assign initial numbers, ... implement IuUP ... well, "just" all that. Currently it looks like we don't need to re-setup the gsmcore. But since that should be rather swift from the ansible, we might as well wipe it once it's on the congress premises (or create a fresh one next to it). I took great care to have everything in the eventgsm git and ansible, even our test VPNs for remote BTSes. ~N -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: From peter at stuge.se Wed Dec 12 16:16:18 2018 From: peter at stuge.se (Peter Stuge) Date: Wed, 12 Dec 2018 16:16:18 +0000 Subject: [osmocom-event-orga] osmoserv online In-Reply-To: <20181212160104.GH1926@my.box> References: <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> <20181210030554.8900.qmail@stuge.se> <20181210114949.GA1991@my.box> <20181210121725.24654.qmail@stuge.se> <20181211003050.GF6861@my.box> <20181212145851.27003.qmail@stuge.se> <20181212160104.GH1926@my.box> Message-ID: <20181212161618.28663.qmail@stuge.se> Neels Hofmeyr wrote: > > disable password authentication > > Absorbed that into the eventgsm.git. Thanks! > Just need to test voice codecs, test the token interaction, dump all sim card > data in the hlr, assign initial numbers, ... implement IuUP ... well, "just" > all that. I think we are in really good shape - better than ever! :) I've asked C3NOC about onsite network configuration but have no response just yet. > Currently it looks like we don't need to re-setup the gsmcore. But since that > should be rather swift from the ansible, we might as well wipe it once it's on > the congress premises (or create a fresh one next to it). > > I took great care to have everything in the eventgsm git and ansible, even our > test VPNs for remote BTSes. Great! We could replicate the setup in a second container and compare. //Peter From peter at stuge.se Fri Dec 14 13:08:20 2018 From: peter at stuge.se (Peter Stuge) Date: Fri, 14 Dec 2018 13:08:20 +0000 Subject: [osmocom-event-orga] osmoserv offline [was: online] In-Reply-To: <20181212145851.27003.qmail@stuge.se> References: <20181205120137.GD5760@my.box> <20181205130643.6927.qmail@stuge.se> <20181208182046.GB16513@my.box> <20181209035427.17214.qmail@stuge.se> <20181210002403.GA28498@my.box> <20181210030554.8900.qmail@stuge.se> <20181210114949.GA1991@my.box> <20181210121725.24654.qmail@stuge.se> <20181211003050.GF6861@my.box> <20181212145851.27003.qmail@stuge.se> Message-ID: <20181214130820.15322.qmail@stuge.se> Peter Stuge wrote: > I'll shut the machine down on Friday for pickup The machine is on its way. I also got a 4-port PCIe NIC and will bring the PCIe riser with me, we don't need ISDN for this event so we can swap out the PCI-X riser. //Peter From lynxis at fe80.eu Mon Dec 24 03:14:40 2018 From: lynxis at fe80.eu (Alexander Couzens) Date: Mon, 24 Dec 2018 04:14:40 +0100 Subject: [osmocom-event-orga] gsm.fe80.eu -> 35c3 gsm server Message-ID: <20181224041440.31c25bd6@lazus.yip> hi, i've changed the ip of the domain gsm.fe80.eu to our new 35c3 gsm hardware. feel free to make yourself comfortable. please deploy the ansible on the new ip. best, lynxis old ip/vm: 80.241.60.93 NEW ip/vm: 94.45.243.151 -- Alexander Couzens mail: lynxis at fe80.eu jabber: lynxis at fe80.eu mobile: +4915123277221 gpg: 390D CF78 8BF9 AA50 4F8F F1E2 C29E 9DA6 A0DF 8604