Re: [DNG] simple-netaid-backend debugged.
Hi Eward, On 7/3/19 8:38, Edward Bartolo via Dng wrote: Hi Everyone, My version of simple-netaid-backend has been debugged to connect when there is only one active wifi hotspot. It was previously failing to connect because there was an error in a while loop which prevented iteration from taking place when there was only one active wifi hotspot. Please, note my graphical frontend does not use unnecessary cosmetics to make it look appealing to the eyes. My aim was simplicity and low use of system processing and memory. Moreover, the backend establishes a connection using low level calls to avoid using ifupdown. It uses instead ifconfig, iwconfig, wpa_supplicant and dhclient. I'm working again on simple-netaid, and i 'd like to share with you the C code for bringing up/down a concrete network interface (void interface_up/down, respectivelly): / Bring up the interface ***/ void interface_up (const char *if_name) { struct ifreq ifr; int skfd = 0; strncpy(ifr.ifr_name, if_name, IFNAMSIZ); /* Create a channel to the NET kernel. */ if((skfd = iw_sockets_open()) < 0) { perror("socket"); return -1; } skfd = socket (AF_INET, SOCK_DGRAM, 0); if (skfd && ioctl(skfd, SIOCGIFFLAGS, &ifr) >= 0) { printf("Activating interface %s", if_name); strncpy(ifr.ifr_name, if_name, IFNAMSIZ); ifr.ifr_flags |= (IFF_UP | IFF_RUNNING); ioctl(skfd, SIOCSIFFLAGS, &ifr); } else { printf("Getting flags for interface %s failed, not activating interface.", if_name); } /* Close the socket. */ iw_sockets_close(skfd); } / Bring down the interface ***/ void interface_down (const char *if_name) { struct ifreq ifr; int skfd = 0; strncpy(ifr.ifr_name, if_name, IFNAMSIZ); /* Create a channel to the NET kernel. */ if((skfd = iw_sockets_open()) < 0) { perror("socket"); return -1; } if (skfd && ioctl(skfd, SIOCGIFFLAGS, &ifr) >= 0) { printf("Taking down interface %s", if_name); strncpy(ifr.ifr_name, if_name, IFNAMSIZ); ifr.ifr_flags &= ~IFF_UP; ioctl(skfd, SIOCSIFFLAGS, &ifr); } else { printf("Getting flags for interface %s failed, not taking down interface.", if_name); } /* Close the socket. */ iw_sockets_close(skfd); } HTH, Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] simple-netaid-backend debugged.
Hi again, On 17/3/19 13:38, aitor_czr wrote: Hi Eward, On 7/3/19 8:38, Edward Bartolo via Dng wrote: Hi Everyone, My version of simple-netaid-backend has been debugged to connect when there is only one active wifi hotspot. It was previously failing to connect because there was an error in a while loop which prevented iteration from taking place when there was only one active wifi hotspot. Please, note my graphical frontend does not use unnecessary cosmetics to make it look appealing to the eyes. My aim was simplicity and low use of system processing and memory. Moreover, the backend establishes a connection using low level calls to avoid using ifupdown. It uses instead ifconfig, iwconfig, wpa_supplicant and dhclient. I'm working again on simple-netaid, and i 'd like to share with you the C code for bringing up/down a concrete network interface (void interface_up/down, respectivelly): / Bring up the interface ***/ void interface_up (const char *if_name) { struct ifreq ifr; int skfd = 0; strncpy(ifr.ifr_name, if_name, IFNAMSIZ); /* Create a channel to the NET kernel. */ if((skfd = iw_sockets_open()) < 0) { perror("socket"); return -1; } skfd = socket (AF_INET, SOCK_DGRAM, 0); if (skfd && ioctl(skfd, SIOCGIFFLAGS, &ifr) >= 0) { printf("Activating interface %s", if_name); strncpy(ifr.ifr_name, if_name, IFNAMSIZ); ifr.ifr_flags |= (IFF_UP | IFF_RUNNING); ioctl(skfd, SIOCSIFFLAGS, &ifr); } else { printf("Getting flags for interface %s failed, not activating interface.", if_name); } /* Close the socket. */ iw_sockets_close(skfd); } / Bring down the interface ***/ void interface_down (const char *if_name) { struct ifreq ifr; int skfd = 0; strncpy(ifr.ifr_name, if_name, IFNAMSIZ); /* Create a channel to the NET kernel. */ if((skfd = iw_sockets_open()) < 0) { perror("socket"); return -1; } if (skfd && ioctl(skfd, SIOCGIFFLAGS, &ifr) >= 0) { printf("Taking down interface %s", if_name); strncpy(ifr.ifr_name, if_name, IFNAMSIZ); ifr.ifr_flags &= ~IFF_UP; ioctl(skfd, SIOCSIFFLAGS, &ifr); } else { printf("Getting flags for interface %s failed, not taking down interface.", if_name); } /* Close the socket. */ iw_sockets_close(skfd); } HTH, Aitor. I'm also working on an alternative to poettering's ifplugd for the automatically wired connect option of simple-netaid. Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] simple-netaid-backend debugged.
> I'm also working on an alternative to poettering's ifplugd for the > automatically wired connect option of simple-netaid. > > Aitor. > Can you borrow code from netplug for that? It does the same as ifplugd. —Tom___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] simple-netaid-backend debugged.
Hi Tom, On 17/3/19 14:38, wirelessd...@gmail.com wrote: I'm also working on an alternative to poettering's ifplugd for the automatically wired connect option of simple-netaid. Aitor. Can you borrow code from netplug for that? It does the same as ifplugd. —Tom I knew netplug, but i tried downloading the sources with *netplugd* instead of *netplug* Thanks :) Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] simple-netaid-backend debugged.
Hi, On 17/3/19 16:04, aitor_czr wrote: Hi Tom, On 17/3/19 14:38, wirelessd...@gmail.com wrote: I'm also working on an alternative to poettering's ifplugd for the automatically wired connect option of simple-netaid. Aitor. Can you borrow code from netplug for that? It does the same as ifplugd. —Tom I knew netplug, but i tried downloading the sources with *netplugd* instead of *netplug* Thanks :) Aitor. Netplug takes a lot of code taken from ifplugd, and i think it should be a much easier way for that. Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] simple-netaid-backend debugged.
On 17/3/19 16:16, aitor_czr wrote: [...] takes a lot of code taken from [...] while it may appear reduntant :) ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] simple-netaid-backend debugged.
Hi, On 17/3/19 13:38, aitor_czr wrote: Hi Eward, On 7/3/19 8:38, Edward Bartolo via Dng wrote: Hi Everyone, My version of simple-netaid-backend has been debugged to connect when there is only one active wifi hotspot. It was previously failing to connect because there was an error in a while loop which prevented iteration from taking place when there was only one active wifi hotspot. Please, note my graphical frontend does not use unnecessary cosmetics to make it look appealing to the eyes. My aim was simplicity and low use of system processing and memory. Moreover, the backend establishes a connection using low level calls to avoid using ifupdown. It uses instead ifconfig, iwconfig, wpa_supplicant and dhclient. I'm working again on simple-netaid, and i 'd like to share with you the C code for bringing up/down a concrete network interface (void interface_up/down, respectivelly): / Bring up the interface ***/ void interface_up (const char *if_name) { struct ifreq ifr; int skfd = 0; strncpy(ifr.ifr_name, if_name, IFNAMSIZ); /* Create a channel to the NET kernel. */ if((skfd = iw_sockets_open()) < 0) { perror("socket"); return -1; } skfd = socket (AF_INET, SOCK_DGRAM, 0); if (skfd && ioctl(skfd, SIOCGIFFLAGS, &ifr) >= 0) { printf("Activating interface %s", if_name); strncpy(ifr.ifr_name, if_name, IFNAMSIZ); ifr.ifr_flags |= (IFF_UP | IFF_RUNNING); ioctl(skfd, SIOCSIFFLAGS, &ifr); } else { printf("Getting flags for interface %s failed, not activating interface.", if_name); } /* Close the socket. */ iw_sockets_close(skfd); } / Bring down the interface ***/ void interface_down (const char *if_name) { struct ifreq ifr; int skfd = 0; strncpy(ifr.ifr_name, if_name, IFNAMSIZ); /* Create a channel to the NET kernel. */ if((skfd = iw_sockets_open()) < 0) { perror("socket"); return -1; } if (skfd && ioctl(skfd, SIOCGIFFLAGS, &ifr) >= 0) { printf("Taking down interface %s", if_name); strncpy(ifr.ifr_name, if_name, IFNAMSIZ); ifr.ifr_flags &= ~IFF_UP; ioctl(skfd, SIOCSIFFLAGS, &ifr); } else { printf("Getting flags for interface %s failed, not taking down interface.", if_name); } /* Close the socket. */ iw_sockets_close(skfd); } HTH, Aitor. I took the code from debian-installer, more concretly from the netcfg udeb package. Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] new freedesktop "standard": /etc/machine-id
On 08.03.19 14:23, KatolaZ wrote: > this is currently managed by eudev in devuan and, IIRC, it is simply > regenerated as a random ID at each boot. I guess it's still there > because it is used by several things, including > session-management-related stuff. We had a discussion on IRC with Mark > (LeePen) about that several weeks ago, and IIRC we concluded that > keeping it around but re-generating it at boot was the way to avoid > breakage. > > Any thoughts on the matter are appreciated, but concrete insight on > the ins and outs are much more useful I guess (read: please let's > avoid a useless uninformed flame about that :P). I would assume that the machine-id is used for: 1. Software Licencing 2. Identifying machines in a pool/swarm/cluster/cloud of machines where static IP adresses are not necessarily a given similar to HP/HP-UX: /bin/uname -i IBM/AIX: /bin/uname -m SGI/IRIX: /sbin/sysinfo -s Sun/Solaris: /usr/ucb/hostid machine-id should theoretical be a serial number or UUID Cheers Mike ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
[DNG] Quirks with Xorg and ASCII (startx and transparency)
Hi all, So, I have just upgraded my machine from Jessie to Ascii, and I am having some issues. Specifically: 1. startx as non-root user no longer works. Now I keep getting "permission denied" for X when it attempts to start. This isn't completely surprising, as I read the upgrade notes, and it said that there have been some changes to xorg, and to get the old behaviour you need "xserver-xorg-legacy" installed. However I have that package installed, but it still doesn't work. What do I need to do to get this working again? 2. Pseudo-transparency on rxvt no longer works. On jessie (and much before), I had pseudo-transparency set using the "-tr" on rxvt. However on ascii rather than getting the background image, I just have a black bg image. Attempts to get this working so far have not succeeded. Anyone come across this? Any ideas how to fix? Thanks! ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Desktop integration
On 14.03.19 10:10, Hendrik Boom wrote: > I get it. They want everything to be easily accessed via mouse with > helpfull instruction in the moment all the way. And that includes > adjusting how their system works. > > The closest I've seen anything come to that is the configurator for the > Linux kernel when you compile your own. BTW: few days ago, I moved out Kconfig into a separate package. I intend to use it in several other places, eg. kernel metaconfiguration (create kconfig's on a much higher level), image building tools (somewhat like ptxdist for Debian ;-)). Maybe that's something that could be interesting for this usecase. --mtx -- Enrico Weigelt, metux IT consult Free software and Linux embedded engineering i...@metux.net -- +49-151-27565287 ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] logging uses of machine-id
On Thu, 14 Mar 2019, goli...@dyne.org wrote: > On 2019-03-14 03:48, Rick Moen wrote: > > Quoting KatolaZ (kato...@freaknet.org): > > > > > I just wonder whether your nice solution is enough for a wider > > > audience used to have things popping up around all the time. I guess > > > it's not, as much as setnet is not a solution to manage networks > > > palatable to a wider audience, as much as apt-get is not the tool used > > > by the large majority of users to install packages, and so on. > > > > > > Tinkerers will always be fine. But a distro that aims to be universal > > > like Devuan must also cater for those who are not willing to tinker > > > around, IMHO. > > > > I'm quoting the above in order to express appreciation for it -- in > > context. Like Steve Litt, I favour the smallest possible entanglement > > with Freedesktop.org 'desktop' components and their characteristic > > tangled dependency trees and difficult-to-justify complications, > > -=but=- the key fact is that Devuan Project has committed to be a > > universal operating system, construed as including DE-style software > > integration. > > > > I would not have supported that commitment, personally, but nonetheless > > can summon the wisdom to avoid expecting this project to abandon its > > principles just because I don't share some of them. And I thank you for > > aptly restating those principles. > > > > > When I first started using Debian, there was discussion of what > "universal" meant. Vocal old timers insisted it referred to the > variety of architectures that Debian would run on not different user > preferences. I am not arguing here; just providing some context of > the definition that has stuck in my mind (which may or may not be > "correct"). I prefer to avoid the term "universal", whose adoption hints on a limited and perhaps authoritarian attitude, at least referring to the meaning of the term in philosophy. Also it makes me wonder what people think of, when they go around making a "universal system" with a "apt-get install anarchy" t-shirt... really?!. IMHO Devuan should aim to be a *base system* (and Debian too!). Being a "base" entails being minimal ("need to work" principle) and reliable for other derivatives, not a moving target, not an opinionated advanced system. Along this line, we are good with providing simple formulas for desktop/embedded/vm system that provide bases for derivatives, as well an efficient (and soon well documented) SDK for making them. ciao ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
[DNG] pmount with exfat
Upgraded a test system running Ascii with Xfce4 to Beowulf today. I noticed that the old system had consolekit and udisks2/gvfs installed and the upgrade for udisks2 in beowulf wanted to install elogind. In this case I decided to ditch udisks2 and try out pmount, but I’ve noticed that pmount doesn’t like exfat filesystems. I can see the bug report on Debian and Ubuntu for pmount has a patch sitting for years unapplied. Has anyone seen a workaround for this, or will I have to run mount as root manually for exfat formatted USB sticks? This system is running fine with consolekit2 so I don’t feel like switching to elogind and installing udisks2 again. Thanks —Tom ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] logging uses of machine-id
On 2019-03-17 6:05 p.m., Jaromil wrote: On Thu, 14 Mar 2019, goli...@dyne.org wrote: On 2019-03-14 03:48, Rick Moen wrote: Quoting KatolaZ (kato...@freaknet.org): I just wonder whether your nice solution is enough for a wider audience used to have things popping up around all the time. I guess it's not, as much as setnet is not a solution to manage networks palatable to a wider audience, as much as apt-get is not the tool used by the large majority of users to install packages, and so on. Tinkerers will always be fine. But a distro that aims to be universal like Devuan must also cater for those who are not willing to tinker around, IMHO. I'm quoting the above in order to express appreciation for it -- in context. Like Steve Litt, I favour the smallest possible entanglement with Freedesktop.org 'desktop' components and their characteristic tangled dependency trees and difficult-to-justify complications, -=but=- the key fact is that Devuan Project has committed to be a universal operating system, construed as including DE-style software integration. I would not have supported that commitment, personally, but nonetheless can summon the wisdom to avoid expecting this project to abandon its principles just because I don't share some of them. And I thank you for aptly restating those principles. When I first started using Debian, there was discussion of what "universal" meant. Vocal old timers insisted it referred to the variety of architectures that Debian would run on not different user preferences. I am not arguing here; just providing some context of the definition that has stuck in my mind (which may or may not be "correct"). I prefer to avoid the term "universal", whose adoption hints on a limited and perhaps authoritarian attitude, at least referring to the meaning of the term in philosophy. Also it makes me wonder what people think of, when they go around making a "universal system" with a "apt-get install anarchy" t-shirt... really?!. IMHO Devuan should aim to be a *base system* (and Debian too!). Being a "base" entails being minimal ("need to work" principle) and reliable for other derivatives, not a moving target, not an opinionated advanced system. Along this line, we are good with providing simple formulas for desktop/embedded/vm system that provide bases for derivatives, as well an efficient (and soon well documented) SDK for making them. I think the current Debian based, Devuan installer for most people is not far from that. It really is a matter of knowing where to bail out of the install procedure to get what you want as a starting point. Maybe I'm missing something and the existing base install is too much for some, but it looks to me as if the jumping off points may just need better definition or maybe documentation. Clarke ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Quirks with Xorg and ASCII (startx and transparency)
On Sun, Mar 10, 2019 at 11:42:48AM +, Ognen B via Dng wrote: > Hi all, > > So, I have just upgraded my machine from Jessie to Ascii, and I am having > some issues. Specifically: > > 1. startx as non-root user no longer works. Now I keep getting "permission > denied" for X when it attempts to start. > > This isn't completely surprising, as I read the upgrade notes, and it said > that there have been some changes to xorg, and to get the old behaviour you > need "xserver-xorg-legacy" installed. > > However I have that package installed, but it still doesn't work. What do I > need to do to get this working again? Hi Ognen, please read through the ASCII Release notes: https://files.devuan.org/devuan_ascii/Release_notes.txt The relevant section is "Starting X from a terminal". You should add a line in Xwrapper.config for that to work. HTH KatolaZ -- [ ~.,_ Enzo Nicosia aka KatolaZ - Devuan -- Freaknet Medialab ] [ "+. katolaz [at] freaknet.org --- katolaz [at] yahoo.it ] [ @) http://kalos.mine.nu --- Devuan GNU + Linux User ] [ @@) http://maths.qmul.ac.uk/~vnicosia -- GPG: 0B5F062F ] [ (@@@) Twitter: @KatolaZ - skype: katolaz -- github: KatolaZ ] signature.asc Description: PGP signature ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
[DNG] Running Devuan Ascii on AWS
I’m looking at migrating some servers from a local ESXi onto AWS, or just rebuilding them from scratch inside AWS. Has anyone successfully run Devuan ascii on AWS? Is there anything special that needs to be installed, like open-vm-tools in VMware? Any other considerations to take into account? Thanks —Tom ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng