21[00:37:23] <H-var> super mega million times better!!! and there are colors! The colors are so beautiful! OMG!!! It's the most wonderful upgrade in the universe a humen being can ever desire!!! Once you go bulleye you never go back
48[00:55:36] <karlpinc> mutante: I use dia. Unfortunately, it does not seem to be getting upgraded to python3. At least not last I looked. So its days are numbered.
49[00:56:17] <karlpinc> mutante: Inkscape works. But, "eh". Not really so good for ER diagrams, which is what I want.
50[00:56:39] *** Quits: badsektor (~badsektor@replaced-ip) (Remote host closed the connection)
51[00:57:03] <mutante> thanks karlpinc and sney. ah... python3-only on bullseye make people notice all the python2-only things :(
52[00:57:11] <mutante> but it was time
53[00:57:15] <karlpinc> mutante: I think that inkscape will actually work for er diagrams in bullseye. Something was missing in buster, but I forget exactly what. And still, kinda "eh".
64[01:08:05] <rockworld> hi, can we use a aliases on network card to create a new ip, let say we have eth0 as interface name and I dont want to touch default setting can I add eth0:1 eth0:2 eth0:3 to add 3 different ip ?
77[01:26:57] <H-var> ok, so here's something that breaks my brain: so I have a mouse, and my mouse re-connects in a split-second after going to sleep both on linux, and the laptop windows systems, and then I have a keyboard, and the keyboard re-connects also in a blink of an eye on that same windows laptop, but on linux it takes more than 20 seconds to re-connect. So what is it? How can it make any sense???
161[02:14:26] <aphysically> hi all, I'm trying to work on making some packaging changes for gentoo
162[02:15:00] <aphysically> my understanding is in debian's packaging of libglslang-dev some downstream changes are made that fix some ffmpeg stuff, and I'd like to bring those changes over to gentoo, but I'm not sure where I can find the downstream patches in debian
163[02:15:11] <aphysically> can someone point me in the right direction for where patch sets are kept?
164[02:15:46] <aphysically> sorry the package is glslang-dev
165[02:18:33] <abrotman> Did you try to grab the source package?
242[04:23:49] <brizz> say I have a line, that is city, state, country, but the line could be missing city or state, or both. Is the only way to 'unpack' the string is to use if len(2): state, country = location, or to use try: except ValueError: ..try: except ValueError under exception? ...seems really dirty. so wondering if there is a better looking way to do this
243[04:24:26] <brizz> oh, i am splitting the line by comma
244[04:24:35] <brizz> example line: menlo park, california, united states
245[04:25:15] <brizz> another example line could be: united states, or: indiana, united states
282[05:05:27] <rockworld> on a dedicated machine I have a block of ip that I need to add in the network interface, do I need to add ip per ip or maybe I can add the entire block one shot
299[05:26:47] <Juerd> rockworld: Per individual IP. A computer doesn't typically get multiple addresses, that's an exception to the rule (a common exception, but still not the way networks were designed)
300[05:27:39] <rockworld> Juerd: alright thank you :)
301[05:28:21] <Juerd> rockworld: A "block" doesn't technically exist. A subnet exists, but may not be exactly the range you got assigned, and in a subnet there are typically (but not necessarily) other addresses that aren't your computer, like a router, the network address, and a broadcast address.
305[05:31:52] <rockworld> Juerd: thats good to know I need some reflexion now
306[05:34:23] <rockworld> Juerd: hum 255.255.255.0 is my server netmask but Im watching a tutorial about that and the guy use different netmask for ipfailover then the server netmask
307[05:34:52] <Juerd> Use the network mask the network administrator has told you to use, and nothing else
308[05:35:41] <rockworld> Im going to try this thank you for the tips I really appreciate
369[06:49:05] <somiaj> warsoul: It varies, though at this point in the freeze, the main thing you'll miss (unless you hit an rc-bug) is security support in bullseye. Many are using it as their desktop/home machine. Waiting is probably best (some think the release is 1-2 months away)
370[06:49:29] <somiaj> rockworld: what do you mean by service directory? What is it you are trying to do?
408[08:05:20] <rockworld> well, I added a new ip and a vhost as a website, I can access the website and I dont understand why the ip doesnt resolve to the host but the host resolve to the ip
430[09:02:15] <jolt> rockworld: not sure what you are doing, but serverhost <name> only matches the name sent in the get request. If you make a request to an IP, well, it doesn't have the name in it; therefor you need to configure either a default host that handles that, or add the IP to it.
467[09:48:14] <supernaut> What is required to be set up before booting from a root on lvm and luks? I've already added the partition to crypttab and installed cryptsetup. However kernel still doesn't ask about the passphrase.
468[09:50:23] <ratrace> supernaut: and cryptsetup-initramfs? I don't remember if it's pulled in automatically
470[09:51:18] <supernaut> ratrace: I'll chroot and check whether it is installed
471[09:51:19] <ratrace> I also don't remember if that bug was fixed, but if it's still not, you'll need "initramfs" in the crypttab options, next to "luks" (last column, comma separated options like in fstab)
472[09:51:48] <ratrace> (update-initramfs -u -a after you're done with changes to it)
479[09:58:09] *** Quits: Tobbi (~Tobbi@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
480[09:58:32] <supernaut> ratrace: weird, cryptsetup-initramfs haven't been pulled with cryptsetup. Im almost sure that the last time it worked just by pulling crypttab.
483[10:01:36] <ratrace> supernaut: now that you mention it, it's been like that since buster release I think, and in fact if you don't have crypsetup-initramfs installed, you get a warning on each initramfs update
484[10:01:55] <ratrace> (if it detects cryptsetup but not cryptsetup-initramfs present)
485[10:02:47] <supernaut> i haven't seen any warning
486[10:02:48] <ratrace> or is it the other way around ... crypsetup-initramfs present, but no crypttab setup ... bah
487[10:03:06] <ratrace> yeah it's the other way around, sorry
651[13:46:15] *** Quits: Nokaji (~Nokaji@replaced-ip) (Quit: "... when the freedom they wished for most was freedom from responsibility then Athens ceased to be free and was never free again.” ~ Edward Gibbon (1737-1794) - Decline and Fall of the Roman Empire, 1909)
690[14:20:26] <Dude-Meister> I'm running Debian Buster stable, and I'm learning as I go. I've stated using SAS drives in LSI Based controllers (I have a Dell PERC H310 that's been cross-flashed for IT mode) I now have a second controller (Perc H200) that I'm tuying with the idea of runing it RAID 0 instead of IT more, becuase I have 4 2TB Seagates, and a couple 4TB Hitachi UltraStar drives I use as offline blackups.
691[14:20:35] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
692[14:22:54] <Dude-Meister> Is there any compelling reason not to create a couple RAID 0 stripe pairs in the Seagates to make backing up two drives at a time a bit simpler and maybe even a bit faster reads and writes in day to day use?and backing
693[14:23:34] <Dude-Meister> I've never run a RAID, so, this is sorta new to me.
694[14:23:56] <Dude-Meister> (As is Linux, really - srill)
695[14:24:35] <qman__> raid 0 essentially halves your reliability
696[14:24:42] <qman__> on a pair of disks
697[14:24:49] <Dude-Meister> Yes, and I'm fine with that.
698[14:25:51] <qman__> you apparently haven't lost too much data before, but you do you
699[14:26:28] <Dude-Meister> 4 bits here at the same tiem as 4 bits there - essentiall (nearly) doubles through-put. I understand that part, adn I know I'll have to blitz two drives and format them as an array, but I don't know if this is fully supported by the card, in Debian
700[14:27:00] *** Quits: clf59 (~clf59@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
701[14:27:03] <Dude-Meister> qman__, you may have missed the part where I said I back up and have redundant copies.
702[14:27:38] <sigint> Dude-Meister, also do your homework on hardware RAID vs software RAID. From your message it's not clear to me which way you're trying to go
703[14:27:49] <Dude-Meister> 3 copies, two in my home and one "Offsite," though someone less recently updated.
704[14:28:05] <Dude-Meister> Dell PERC H200 card
705[14:29:11] <Dude-Meister> The Dell H310 is running IT mode, the Dell H200 I just purchased, adn stopped befor flaching it to IT and said, "Wait, is it possible I can run RAID 0 and LIKE IT? So, here I am.
706[14:30:18] <jelly> Dude-Meister, one uses raid0 when one needs to squeeze out iops performance. Perfomance on large SATA disks is horrible anyway, and 2xhorrible = still horrible. Go with concatenated disks instead of raid0 if you need the capacity, or better, use them separately and lvm or zfs on top.
707[14:30:33] <Dude-Meister> I JUST backed up 2 of the 4 2TB SAS drives, so i COULD blitx them - RAID 0 them on the new card, then simply transfer the data back.
708[14:31:34] <Dude-Meister> (I have a second identical machine, as when I speced a new machine, I built two - becasue - redundancy adn identical hardware is - more simple - and more like the real pros.
709[14:32:28] <Dude-Meister> Same Asus boards, same Ivy Bridge unlocked CPUs (NEVER overclocked adn stable as HELL)
711[14:33:25] <Dude-Meister> Same "Samsung" "Super-ram" that gammers were oc-ing the heck out of - NEVER overclocked in these machines - and so forth.
712[14:33:47] *** Quits: wintersky (uid453465@replaced-ip) (Quit: Connection closed for inactivity)
713[14:34:04] <Dude-Meister> And no, so far, I have never lost important data, and have been using computers as a "home user" since 1980.
714[14:34:18] <ratrace> there's always first time, but if you have backups, all good
715[14:34:49] <Dude-Meister> (I take that back. The first machine I was allowed to use (as a kid) was a TRS-80 Model III, with a cassette tape
716[14:35:01] <ratrace> btw raid0 is not going to be a speed improvement you'd expect in most cases.
717[14:35:09] <Dude-Meister> Some of the tape were - tapes
718[14:35:14] <Dude-Meister> :/
719[14:35:27] <jelly> tapes are good storage media.
720[14:35:45] <jelly> proper humidity and temp, much better than disks
721[14:35:54] <sigint> jelly, LTO tapes, not cassette tapes
722[14:36:02] <jelly> (and lack of dust)
723[14:36:32] <Dude-Meister> No 10" floppy, like the TRS-80 model II, adn the fist single density single sided disk in the Model 3 was $999 PLUST the $380 PSU for the drive.
739[14:40:01] <jelly> hw raid is nice if you have battery or flash backed write cache
740[14:40:26] <ratrace> hw raid stopped to be nice long time ago
741[14:40:29] <Dude-Meister> jelly, this is making sense to me, only because there are SO MANY other people ding things this way that I am likely to get help more easilly if I have a question.
749[14:42:45] <Dude-Meister> See? I'm asking a simple querstion about my rig - I provided the info about wheat hardware I have - and it's sort of - well - who likes which way more. I get it, but, can you see how it's cumbersome for a poor boy that DREAMED of being able to AFFORD the cards i NOW HAVE - (living off depreciation) and CAN learn with?
750[14:43:38] <Dude-Meister> I only have 16 GB or ram in these boxes, adn don't have moey to throw at it, or I'd have a Dell R710
758[14:46:20] <ratrace> well, if you're short on gold coins, then hw raid and brand name chassies is the last thing you'd want. contrary to what the marketing depts of Dells and IBMs and HPs would like us believe ... one can build equally reliable machines without the brand tax.
759[14:46:44] <Dude-Meister> Now, the 4 2TB Seagate SAS drives I now own, (had to slowly save for THOSE, and add to the stable a little at a time) contain pretty much (Soup to Nuts) including personal finance and family pic and vids, plus "current work stuff."
760[14:47:23] <Dude-Meister> Those that live off depreciation can. :)
761[14:48:01] <Dude-Meister> Solid powersupplies, shopped hard adn puchased wisely - AND knowing how when and WHY to "Re-Cap"
762[14:48:08] <jelly> if you can translate "fun" to "specific technical requirements" then a tech channel can offer sugestions
763[14:48:11] <Dude-Meister> MoBo's are all Mil Spec Caps
765[14:49:13] <Dude-Meister> I just want to know If I can run the small 2TB drives in a hardware RAID in Bebian using this PERC H200 - using 8 lanes.
766[14:49:49] <jelly> yes you can, linux supports mptsas* controllers well
767[14:50:01] <Dude-Meister> I already run an H310 HBA, and it runs fine, as I've added additional cooling becasue it's not in a server chasis, so i don't want to overheat and burn up the LSI Chipset.
768[14:50:22] <jelly> it probably has temp sensors
769[14:50:39] <jelly> (but might or might not require vendor tools to read them)
770[14:51:13] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
771[14:51:28] <jelly> LSI hw raid CLI tools have some of the worst syntax I've ever seen
774[14:52:01] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
775[14:52:10] <jelly> seriously, look up MegaCLI syntax. Or better, don't if you want to have lunch today and keep it
776[14:52:13] <Dude-Meister> Adding a fan to the heat-sink is VERY wise when runnign these cards, becasue desktop chasis - (Even my custom Lian-Li box with 5 5-1/4 external andn 10 internal 3.5" bays and 6 big fans is - not really flowing like a REEAL server chasis, so . . . yeah . . . that. Add a fan to LSI based HBA's
779[14:53:02] <ratrace> jelly: so that's the "nice" part about hw raid you mentioned earlier? :)
780[14:53:45] <sigint> Dude-Meister, server chasis have terrible airflow, they need to use high RPM fans because their cannot fit bigger fans. Your LSI card is probably happier in a tower with decent airflow
781[14:53:51] <Dude-Meister> I would like to leave time for lunch, and would just as soon not walk right into an engineering problem that's already well explored by others. :)
782[14:54:18] <Dude-Meister> sigint, it's - not where the fans are.
783[14:55:06] <Dude-Meister> All the fans are really on the drives, therefore it's a dead spot where the card lives, so, adding a littel fan DID significantly lower temps.
784[14:55:10] <Dude-Meister> Heat kills.
785[14:55:33] <Dude-Meister> I'm not about loosing my daughter's graduation video! :)
787[14:56:22] <sigint> I get it, I'm just responding to the "additionnal cooling because it's not in a server chasis"
788[14:56:31] <Dude-Meister> I just though it would be cool to have a littel more speed from these littel SAS disks, but not if it's going to cause headaches.
791[14:58:01] <Dude-Meister> Engineers design servers to "Not fail" is possible, so - kinda actually do build them with airflow paths taken into account, probably moreso that the one's building cases with lights and "windows"
795[14:58:37] <ratrace> the high RPM thing is due to component density but they're designed as much as possible with an actual flow
796[14:58:38] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
797[14:59:20] <ratrace> that said, "desktop" chassies require less effort to achieve the same flow, at the expense of much smaller density and much more "wasted" space
798[14:59:22] <Dude-Meister> In any case, I think I've gotten my answer as to whether I am going to continue forward with flashing this shiny brand-used card inso IT mode. :)
806[15:00:15] <sigint> ratrace, yes that's my point. Airflow wise, a desktop tower is actually much better because it's not optimized for density
807[15:01:57] <jelly> ratrace, no, that's the part about "I wouldn't touch LSI with a 10 foot pole, given any other option"
808[15:02:09] <Dude-Meister> Not "All" desktop cases are better engineered, adn in reality, not all server chasis are poorly engineered with HBA's failing non-stop either. It depends on how much $$$$ the design teams have in their budget.
810[15:02:41] <jelly> ratrace, but people use them all the time in HBA mode for 8-12-16 disks
811[15:02:50] <jelly> 'swhat they're good for
812[15:03:05] <ratrace> sigint: actually I disagree. took me some effort to design my worskstation/server tower for optimal flow precisely becuse the fans are typically low pressure
813[15:03:14] <Dude-Meister> Anyhow - fine friends - thank you for sharing your knowledge with me.
815[15:03:32] <jelly> Dude-Meister, what does "IT" mode mean btw?
816[15:03:59] <sigint> ratrace, I'm not sure I follow, fans with low static pressure are fans with great airflow
817[15:04:26] <ratrace> sigint: and the damned modern gpus and cpu coolers are designed for NO Flow at all. gpus take internal aire and dissipate internall (despite them having access from behind fro some cool air suckage)
818[15:04:27] <sigint> You cannot optimize for both
819[15:04:58] <ratrace> the cpu ones just cavitate and swirl teh same air around. so you need powerful bottom -> top flow which is not typically done
820[15:05:05] <sigint> Exactly, CPU and GPU get high static pressure, case fans get low static pressure but great airflow
821[15:05:51] <sigint> going back to the server chasis topic, server chasis don't have this luxury of matching fans this way
823[15:06:04] <ratrace> I found some neat combos with watercooled cpu so its fans are the entire top row which also suck air upwards and comform with natural hot air wnts to raise, flow. the only "problem" is gpu that still swirls the air inside but I "fixed" that by adding front row to create positive pressure inside the chassis
868[15:53:40] <dpkg> <km> dkr: if a web using port 8080, so to avoid haveing to type the port in the url i to change the browser default port 80 to 8080? <twb> km: you couldn't perform enough fellatio in a decade to get that changed
903[16:23:39] <Dude-Meister> Okay, I'm not sure what I'm seeing. I went ahead that plugged that Dep Perk H200 in, just to see if perhaps it was one that has had IT mode firmware on it from day 1 - because as I was pulling info and the "IT Mode" .bin, I saw something suggesting that the H200 may have commonly shipped already running in IT mode
904[16:24:28] <Dude-Meister> Put a "unimportant" (Scrub pile) little disk on it, and booted.
906[16:26:19] <Dude-Meister> I got a 80 columb screen while the card's firmware booted (Which I do not get on the H310 that I had to flash, because it was in fact running IR FW) and bingo - Debian boots and I can see this old 320 GB disk formatted either NTFS or EXT4 (I'm not sure which)
907[16:28:08] <sney> is it a /dev/sdX device? if you do 'fdisk -l /dev/sdX' as root, does Disk model: say the disk's actual model, or something related to your perc?
908[16:28:20] <Dude-Meister> The 80 column screen that the card's firmware loads before the system boots offer's a disk manager by pressing "?" key - and I think it was Dell branded Firmware.
912[16:30:35] <Dude-Meister> It shows it's proper lable in "Disks" and in Nemo.
913[16:30:45] <Dude-Meister> sney, ^^^^^^
914[16:30:47] <sney> if you answer the actual questions
915[16:31:23] <Dude-Meister> I was just checking, and also about to ask if I should use just sdc in that fdisk call?
916[16:31:35] <Dude-Meister> (It shows up in Disks as SDC1
917[16:31:35] <sney> yes, just sdc.
918[16:31:40] <Dude-Meister> KK
919[16:32:37] <Dude-Meister> Yup! It shows the Disk, not the HBA
920[16:32:48] <sney> great, you're in IT mode.
921[16:32:55] <Dude-Meister> Cool!
922[16:33:52] <Dude-Meister> You know, some might flash to LSI firmware to avoid the pause to give the option of loading the Dell Disk Management utility for this card, but, I might LIKE having that at some point.
923[16:34:30] <Dude-Meister> sney, you've helped me understand how to work with multiple disks in Debian before, adn again . . . yeah
924[16:34:38] <sney> it's not getting in your way, and flashing firmware carries a risk of bricking it
925[16:34:54] <sney> LSI HBAs take a while to boot too. that's normal enterprise hardware stuff.
926[16:34:58] <jelly> Dude-Meister, what do "IR" mode and "IT" mode mean?
927[16:35:00] <Dude-Meister> So happy this is all good news. i just got this card, and am as happy as a tornado in a trailer park!
928[16:35:19] <sney> jelly: raid controller on, or passthrough, respectively
929[16:35:33] <jelly> HW RAID or clean HBA, gotcha
930[16:36:04] *** Quits: JohnML (~john1@replaced-ip) (Remote host closed the connection)
932[16:37:27] <Dude-Meister> jelly, So, unless I'm running a software RAID - it seems I can still run JBOD without having to fart around, because I just run as JBOD.
933[16:38:20] <Dude-Meister> Less screwing around and stability as the FIRST order of business is precisely why I do NOT (and have not) runa RAID
934[16:39:54] <Dude-Meister> I was asking if I might want to stripe the drives that have NON CRITICAL data on them, just because A) The are OLD 2TB drives, and B.) They say "Seagate" on them, not HGST Ultrastor, adn are therefore SLIGHTY more prone to failure anyway.
935[16:40:08] <Dude-Meister> Ultrastar*
936[16:40:38] <Dude-Meister> I don't serve OR pull large amouts of data from the internet.
937[16:40:59] <jelly> haviing a jbod is kind of a prequisite to choose running sw raid
942[16:43:20] <Dude-Meister> Yes, but it the IR firmware is running, Software RAID (Whicvh is a layer - and therefore more complicated that needed - by me - AND comes with additional ioverheads on a machine I'd LIKE to be - simple and PDQ BOTH, with stability at the fore-front) If it's running IR, I apparently still use it as JBOD.
943[16:43:44] <Dude-Meister> It's inly RAID SOFTWARE that loosed it's mind if the IR firmware is present.
944[16:44:09] <Dude-Meister> I'm NOT GOING TO RUN - software RAID - period.
946[16:46:04] <Dude-Meister> If the chipset on the RAID card is not capable of keeping load off of whe CPUs i have, (Which are already older than "The latest (Ivy Bridge "K" era, so 4th gen?)
947[16:47:26] <Dude-Meister> I'm not going to re-arrange everything I already own adn replace it with better stuff, because, it's simply not an option.
948[16:47:29] <Dude-Meister> :)
949[16:47:29] <jelly> what load. You have a couple 7200rpm disks.
967[16:51:45] <Dude-Meister> Ok. You may be right - because it's not FreeNAS or unRAID, adn I'm trying to learn about those too, but not right now. I have - you know - tings. :)
969[16:52:23] <Dude-Meister> Like Early Sunglow corn, Rutgers, Red Brandywine and Large Cherry tomatos.
970[16:52:40] <Dude-Meister> But, they don't run Debian, so . . . .hehehehe
971[16:52:59] <jelly> well that's all fine, just that you're aware of the safest practices before choosing a road less traveled
972[16:54:09] <Dude-Meister> I have to pay bills and make food - and WISH I knew all this stuff, but there's only so many hours in a day, adn there's only one "Growing Season" per year. If you miss it, well, no food.
973[16:54:29] <Dude-Meister> Amish is one example of a road less traveled, yes?
975[16:54:57] <Dude-Meister> Well, around where I live - not so much.
976[16:55:08] <sney> !ot
977[16:55:08] <dpkg> #debian is primarily a support channel for Debian users. Please keep the discussions in #debian on-topic and take longer discussions and non-support questions to #debian-offtopic. Imagine the chaos if each of the hundreds of people in the channel felt the need to wander off topic for a few minutes every day.
980[16:56:06] <Dude-Meister> Thanks sney. I was mearly explaining to a software RAID sales staff why i need to jsut do what works right now. RAID 1 - but by hand.
981[16:56:22] <sney> let us know if you actually need help with something
982[16:56:25] <blackroot> by talking about gardening.. lol
983[16:56:25] <Dude-Meister> I don't want two drives spinning all the time.
984[16:57:34] <Dude-Meister> No, blackroot, by giving an example of why I'm gonna use what I found works, and came here to report what the Dell H200's card bios is in fact, likely to work for me.
985[16:58:36] <blackroot> :)
986[16:59:20] <sney> you added a disk controller. now you can have more disks, and/or it's likely faster than your motherboard's onboard sata. hooray! no more needs to be said.
987[16:59:27] <Dude-Meister> And give a strong example of the reason I don't need to be convinced to change my entire rig becazsu e"Software RAID is what I should be doing right now - gently.
988[17:01:17] <Dude-Meister> Prezactilly sney, and now i MIGHT as some other questions, but, since I don't have any, it is my job to clear the channel, and so I shall. Peace all, and again, thank you for the help, sney adn others.
992[17:02:11] <sney> I always forget that that guy needs a little extra crowd control than most /ot
993[17:03:15] <blackroot> sney: i think he is actually doing gardening (weed), and he's so high he THINKS he manages a server. ,y first impression anyway
1002[17:11:20] <jpmh> is there a command line command that I can use to stop the screen blanker blanking my screen after some length oftime.?
1003[17:11:57] <imMute> jpmh: you mean on the text consoles? run setterm -blank 0
1004[17:11:57] <somiaj> xset s off, and xset -dpms if you are talking about the default blanking in xorg
1005[17:12:44] <jpmh> somiaj: Ihave tried those - certainly the latter re-enables the display, but it doesblank again
1006[17:14:52] <imMute> jpmh: are you talking about the text-only virtual consoles, or a GUI desktop?
1007[17:14:57] <somiaj> many DEs will have their own set of power management tools you need to use, I was just talking about the default xorg one in cases you don't have a DE with power management.
1008[17:15:07] <somiaj> though as imMute pointed out, it depends on if you are actually in xorg, or just on a tty
1010[17:17:24] <jpmh> imMute: and somiaj I have an app that has run on raspian for years that does a use of the browser to display a set of slides, being the menu in coffee shops. Since Rapsian is Debian I expected it to work on Debian but the screen blanks
1011[17:18:30] <jpmh> I suspect that the xset commands are being performed for the "wrong" device. The script starts the browser after setting: export DISPLAY=:0.0 - and in favt if I run it manyually fromanotehrr tty then all is fine, the xsets seem to happen
1012[17:18:41] <imMute> jpmh: Raspbian has their own configuration/defaults on top of Debians, so that doesn't really apply.
1014[17:24:40] <jpmh> imMute: yes - but I thought that the xset that I am using on Raspian would work on Debian and it does not, at least does not when the script is invoked by an @reboot line in the crontab
1015[17:25:41] <imMute> jpmh: crontab entries are not executed under the same "context" as your graphical environment, so I'm not really surprised...
1016[17:26:32] <somiaj> jpmh: xset works on debian just fine, the issue is something else as being pointed out.
1017[17:27:00] <somiaj> I use it regurally to stop my screen from blanking
1018[17:27:23] <somiaj> though depending on your needs you could write an xorg.conf snippet and just disable this feature when xorg is loaded
1019[17:27:43] <somiaj> or you could write a custom .xsession that sets up your x-enviorment the way youw ant it
1021[17:30:31] <jpmh> somiaj: as I say, there seems to be some issue that I am not understanding then. When my script which includes the xset commands is run by me at the shell, all works fine. When it is in the @reboot it does not - the browser starts fine, just the blank still happens
1022[17:31:00] <imMute> jpmh: "at a shell" - does that mean in a tty? in a terminal emulator running under your graphical desktop?
1024[17:32:18] <jpmh> imMute: what I mean is that if I ssh into the machine and manually run the scriptat the shellit works as expected - but when the scriptis started from the @reboot it does not
1025[17:32:20] <somiaj> I think @reboot is probably a poor way to realiably get an xsession up and running
1026[17:32:34] <jpmh> maybe the issue has NOTHING to do with xset but my undertanding
1027[17:33:20] <jpmh> somiaj: the xsession is there because the OS has it anyway - I am merely using the @reboot to start something that sets DISPLAY=:0.0 on that monitor
1028[17:33:54] <imMute> jpmh: you can't use @reboot in crontab to set environment variables for *anything* outside of that crontab entry...
1029[17:33:58] <jpmh> my real confusion here is that this has been working and continues to work find on Raspbian
1032[17:34:19] <somiaj> and what makes you sure that the xsession is even up and running by the time the @reboot option is run, this is a poor solution to your problem.
1033[17:34:35] <jpmh> imMute: my crontab entry is merely @reboot startIt.sh
1035[17:35:04] <jpmh> imMute: the startIt.sh script is what does the DISPLAY=, xset, and then starts the brwser
1036[17:35:05] <somiaj> Do you not even use a full path to startIt.sh?
1037[17:35:06] <imMute> jpmh: assuming startIt.sh starts that browser thing, you really should be using systemd services, possibly user scoped ones (I'm unfamiliar with how that stuff works on graphical desktops)
1039[17:35:41] <somiaj> personally I would just setup a proper .xsession that runs and configures xorg like you want it to when it boots (assuming you have some auto login setup)
1040[17:35:43] <jpmh> somiaj: yes, I do use the fullpath, and as I say, I knowthe scrit is running since sure enough I get the slide show
1041[17:35:51] <somiaj> and depending on your DE just setup the autorun apps correctly.
1042[17:36:41] <jpmh> somiaj: I am not against making the changes that you suggest - however, I am nt a fan of making changes to avoid an issue that I don't understand
1043[17:36:58] <somiaj> the issue is cron is a bad tool to setup your enviroment/xsession from
1044[17:37:03] <jpmh> the ice thing about the @reboot method has been that I can test the script at the shell
1045[17:37:24] <imMute> jpmh: but do you understand how it *does* work on the other system?
1046[17:38:10] <jpmh> somiaj: OK - and why is it a bad choice - it seems to set the export DISPLAY=:0.0 correctly, otherwise I would be seeing nothing, what is it that is not being set with xset
1047[17:38:47] <jpmh> imMute: don't undertand your question
1048[17:39:34] <jpmh> realistically, anyway, it seems silly to disable a feature with xset that I did not want on in the first place. Is there noweher that I can configure it not to bother to blank at all
1049[17:39:54] <imMute> jpmh: you want to make a change to avoid an issue you *dont* understand, but do you even understand how the working system works? like *why* does it work
1051[17:40:39] <imMute> jpmh: if you want to configure every little thing, I suggest you check out LFS. otherwise you'll have to accept that the defaults might not always be exactly what you want.
1054[17:42:08] *** Quits: Haudegen (~quassel@replaced-ip) (Quit: Bin weg.)
1055[17:42:09] <jpmh> imMute: no - I want to undertand - then, I'm happy to do something like a correct auto-start - what I don't underatad is why the DISPLAY= in the script wrks when run from crontab but the xset does not
1056[17:42:29] <somiaj> as said, configure your xsession correctly, there are various ways to do this. It can depend on which DE you are using and how you login.
1057[17:42:57] <imMute> jpmh: I don't know either. but I do know that crontab is the wrong tool for this job. so rather than try to figure out how to make the wrong tool work, I'd personally rather switch to the correct (or at least better) tool for the job.
1058[17:44:30] <somiaj> do you have some delay in your script, because I still see @reboot from cron running before your xsession is fully loaded and ready to go.
1059[17:44:45] <somiaj> But still I just see using cron for this a poor method, as you see it isn't quite working.
1060[17:44:51] <jpmh> imMute: I 100% agree that I should just have an autostart - question is, why will that be any different?
1061[17:45:55] <jpmh> somiaj: great idea - let me try that delay - that would make100% sense - then I would undertand and be happy to do it right
1062[17:46:13] <somiaj> but really cron is a terrible tool for what you are trying to do, learn to use correct tools.
1063[17:46:21] <jpmh> guys, somiaj and imMute thanyou so much for your patience
1064[17:46:44] *** Quits: Filohuhum (~filohuhum@replaced-ip) (Remote host closed the connection)
1065[17:46:48] <jpmh> somiaj: why is cron so bad
1066[17:47:16] <somiaj> It was not designed for what you are using it for.
1067[17:49:04] *** Quits: prirai (~Priyanshu@replaced-ip) (Remote host closed the connection)
1068[17:49:05] <imMute> jpmh: cron is designed to run system-level things periodically. you're using it to run something at startup (at some indeterminate time relative to the startup of the graphical environment) against a logged-in user's graphical desktop
1069[17:49:30] <imMute> jpmh: and on top of that, systemd timers effectively remove the need for a separate cron daemon, and are easier to manage and debug
1070[17:50:21] <imMute> jpmh: for example, running your script from cron and running it from an SSH runs it in two different "contexts" (which is likely why it doesn't work in one of them). systemd timers trigger services. whether a service starts from a timer or from manually being started, the context is [nearly] identical.
1071[17:50:45] <jpmh> imMute: it does have the @reboot - so I would not agree that it is not designed for startup - but I do accept that it may be performing too early
1072[17:50:47] <imMute> (I say nearly, because timing is the one thing that would be different - one starts during boot at some stage, the other happens much later when the human tells it to happen)
1073[17:50:51] <somiaj> in essence you are using cron as a hack for properly configuring the xsession the way you want it.
1080[17:54:22] <imMute> jpmh: if that works, then that means the crontab script was running too early, right? that's something that wouldn't have been a problem with somiaj's xsession solution
1081[17:56:38] <jpmh> imMute: yes - and that would probably make sense - the new machne isprobably booting at a different speed to the Pi, and potenially Debian v Raspian (which is now Raspberry Pi OS)
1083[17:57:42] <jpmh> and I BELIEVE at this time that we have solved it - the screen has not blanked - I was not scientific to time how long, but my gut feel is that it would have blanked by now
1084[17:58:36] <somiaj> Linux often offers various ways you can hack things to work, but just because cron has @reboot, dosen't mean it is the right tool for the job
1085[17:58:56] <somiaj> as for testing, for setting up an xession correctly, all you have to do is 'login' to your xsession to test if it is setup correctly.
1086[18:02:57] <jpmh> somiaj: thosemachines are headless - so logging in is not quite as easy as that
1088[18:03:25] <imMute> jpmh: it's not headless if it has a monitor and a graphical environment...
1089[18:04:17] <jpmh> imMute: well itis not completely headless but it does not have input devices - am I missing something, how would I log in on that session?
1090[18:04:28] <jpmh> I should add that I am NOT a GUI typeguy
1091[18:04:54] <imMute> jpmh: I think there might be a way to get systemd to restart the graphical environment, and I assume you have autologin configured.
1092[18:05:35] <jpmh> I have 1200+ coffee shopps running my point of sale system - it was all programmed with NO GUI - the POS itself provides an interface in a web-browser so the users do inetract with JS, CSS, HTM etc
1093[18:06:07] <imMute> interesting that you consider HTML+CSS+JS not a GUI
1094[18:06:16] <jpmh> I have to say I have managed to this point to avoid needing to undertand systemd -
1121[18:29:43] <somiaj> wanko: found that from the changelog, seems it was added in 1.13.2-1, so should be there from buster onwards (and backports if using stretch)
1122[18:30:00] <wanko> yes i have it downloaded, but nginx still doesn't recognize rtmp somiaj
1125[18:31:01] <somiaj> the changelog seems to suggest it works, there was even some docs and examples added. I don't use it, so unsure if anything special needs to be done, but I doubt you need to recompile if running buster
1133[18:38:18] <somiaj> well to me it appears that it is prebuilt, did you configure nginx correctly? Though maybe more nginx support could help you debug things better.
1134[18:38:53] <somiaj> Since there is a package, it appears it has been prebuilt and you should be able to just use the provided binary packages. Might need to wait for someone with more expreience to give any pointers on getting it to work though
1135[18:38:56] <sney> it looks like nginx in buster uses the same sort of mods-available/mods-enabled symlink style configuration as apache, make sure that symlink is in place and restart the daemon?
1136[18:39:18] <somiaj> oh yea, good point, forgot about that.
1153[18:56:12] <UndrWater> i'm trying to update gutenprint to the latest version, as it holds a driver i need for my printer. while i was able to install gutenprint via source, but it seems it's not working with the installed version of cups
1167[18:59:33] <tomreyn> UndrWater: you don't usually upgrade software within a debian release, but upgrade your debian release to get newer software.
1168[18:59:41] <jelly> UndrWater, which relesae of gutenprint do you need?
1169[18:59:47] <jelly> or, which printer
1170[18:59:58] <UndrWater> the printer is epson artisan 1430
1171[19:00:29] * karlpinc can't recall how to ask judd about whether a self-backport would work.
1172[19:02:17] <UndrWater> i'm currently on buster...how "bleeding edge" is bookworm?
1173[19:02:37] <sney> bookworm hasn't started yet
1174[19:02:56] <UndrWater> so then bullseye?
1175[19:03:11] <sney> bullseye is a few months away from the release. some stuff may be turbulent for bugfixes, and security support is slowish at this part of the cycle, but I'm using it on 2 machines with no issues
1176[19:03:35] <sney> !bullseye freeze
1177[19:03:35] <dpkg> the freeze for Debian 11 'bullseye' has begun on 2021-01-13. replaced-url
1178[19:03:43] <jelly> UndrWater, I see an /usr/share/foomatic/db/source/printer/Epson-Artisan_1430_Series.xml on my buster installation, without any additional software
1187[19:06:22] <dpkg> First, check for a backport on <debian-backports>. If unavailable: 1) Add a deb-src line for sid (not a deb line!); ask me about <deb-src sid> 2) enable debian-backports (see <bdo>) 3) apt update; apt install build-essential; apt build-dep packagename 4) apt -b source packagename 5) dpkg -i packagename-ver.deb To change compilation options, see <package recompile>; for versions newer than sid see <uupdate>.
1188[19:06:46] <tomreyn> so buster should have support for it
1194[19:09:45] <UndrWater> tomreyn: i'm not sure it's the release...i THINK it's compatibility with the current version of CUPS. the driver for the printer is now available, but i'm stuck on "processing since..." when attempting to print.
1199[19:11:19] <UndrWater> how do i get that version
1200[19:11:30] <jelly> that's what comes with buster
1201[19:11:57] <jelly> pulled in by printer-driver-gutenprint
1202[19:12:30] <jelly> kind of looks like support is meant to exist
1203[19:13:39] <UndrWater> yeah...full disclosure, this is raspbian....so maybe a limitation there. it looks like i'll have to dig deeper to find out what's going on
1399[20:16:05] <jhutchins> drl: processors are pretty consistent, but graphics can vary.
1400[20:16:08] <jhutchins> !amd
1401[20:16:09] <dpkg> Advanced Micro Devices is a semiconductor manufacturer. For support of ATI/AMD graphics hardware, ask me about <ati>. See also <amd64>, <coolnquiet>, <k7>.
1403[20:17:03] <jelly> drl, in general, it's safer to pick hardware that's been at least 6, better 12 months on the market if you plan to run debian without many issues
1404[20:17:06] <tomreyn> you may need / want replaced-url
1408[20:18:13] <jelly> drl, also in general, if the models or the whole generation are newer than 6 months you don't even have to bother trying current debian release
1427[20:25:54] <drl> jhutchins, jelly, and tomreyn, thanks. My system is about 8 months old, but I'm having lots of problems trying to use Linuxmint.
1428[20:26:12] *** Quits: xet7 (~xet7@replaced-ip) (Remote host closed the connection)
1429[20:26:13] <karlpinc> (And at this point it might be good to go with bullesye instead of buster, although it won't be as stable or as quick with security updates).
1486[20:52:47] <greycat> Normally if something "suddenly stops working", the two questions we'll ask are: 1) What did you change? 2) What does the log file say?
1506[20:56:46] <Bushcat> greycat, its lot of leg work, cos lot of PHP codes, lot of Python-Django codes are running from Apache as sister server from this machine too
1507[20:57:03] <greycat> then why the hell are you running a tor browser on it
1510[20:57:27] <oxek> Bushcat: you'll need to do that legwork anyway, because debian 9 has run out of some support already
1511[20:57:32] <Bushcat> cos Big Sky, Law and Order Organized Crime, episodes I need to download
1512[20:58:04] <greycat> if it's not a browser then they can run it on one of their half a dozen Debian 10 headless servers instead of this one peculiar Debian 9 server
1513[20:58:06] <jelly> Bushcat, get your warez on a recent Debian.
1621[21:37:57] <NetTerminalGene> Bushcat: keep me informing about your funny hybrid stuff please
1622[21:38:51] <Bushcat> well, with cinnamon in it standalone, I felt problems hence I decided to install other DE
1623[21:39:26] <Bushcat> for example in this machine, where I am writing you, its Debian 9 and I installed all the DE and it never cause issue running programs
1624[21:40:05] *** Quits: chele (~chele@replaced-ip) (Remote host closed the connection)
1627[21:40:39] <Bushcat> i am not a Linux System Admin, but I utilize Linux as BEST Server for Web Applications, I work with few FULL STACKS, PERN Stack, LAMP Stack, Python-Django Stack
1655[21:50:20] <dpkg> #debian is primarily a support channel for Debian users. Please keep the discussions in #debian on-topic and take longer discussions and non-support questions to #debian-offtopic. Imagine the chaos if each of the hundreds of people in the channel felt the need to wander off topic for a few minutes every day.
1656[21:50:33] <oxek> !tell NetTerminalGene about rockylinux
1744[22:09:55] *** Quits: tzf (~tyzef@replaced-ip) (Quit: je quitte, quitte a te dire que qui te dira que j'ai quitté, en quittant sans quittance equitable!)
1792[22:31:23] <BuenGenio> good day gents and lassies
1793[22:32:11] <BuenGenio> hopefully a quick one - for some reason my trusted apt keys / "Trusted software providers" keys, keep randomly disappearing on all my laptops
1834[22:38:33] <oxek> !tell BuenGenio about ubuntu
1835[22:38:39] <BuenGenio> now I know your tolerance levels :)
1836[22:39:22] <BuenGenio> @oxek - I just came from there, can't stand that channel - it's too loud, and they're busy arguining about something trivial atm
1837[22:39:27] <jhutchins> BuenGenio: It's just that advice that would fix Debian can break Ubuntu, and most of us don't run both so we don't know what's changed.
1838[22:40:06] <jhutchins> There's probably a channel on oftc.net, it's probably quieter.
1839[22:40:21] <oxek> BuenGenio: the community is part of the distro you pick. Don't like ubuntu community? Switch to debian.
1853[22:51:01] <BuenGenio> on a related, and ultimately the reason I don't think authentication is something that can simply be overlooked (especially in times when CVE's are rolling out like it's pandemic)... - what available and standard issue tools should I be reading up on and make sure are installed to verify the integrity of files (save configs of course) corresponds to that of original Distro provided debs?
1854[22:51:30] <BuenGenio> I'm not sure how debsums works to tell me files are OK, if it was itself installed without any authentication available
1855[22:51:57] *** Quits: platvoeten (~platvoete@replaced-ip) (Remote host closed the connection)
1856[22:52:20] <H-var> what's "trixie-updates"?
1857[22:52:53] *** phunyguy is now known as phunygguy
1867[22:55:18] <H-var> I have: trixie-proposed-updates, trixie-security, trixie-updates all enabled
1868[22:55:28] <greycat> So you're from the future? What's it like?
1869[22:55:38] <H-var> I installed cinnamon from debian.org
1870[22:55:42] *** Quits: polymorphisme1 (~Thunderbi@replaced-ip) (Remote host closed the connection)
1871[22:55:48] <H-var> and it has a built-in upodater, like kde does
1872[22:56:29] <H-var> automatic updater program
1873[22:56:35] *** Quits: Jerrynicki (~niklas@replaced-ip) (Remote host closed the connection)
1874[22:56:49] <BuenGenio> ... i tried tiger, samhain and aide, but they all rely of the assumption that there is a "clean" reference state you can do the initial scans, but I'm simply way more paranoid than assuming my system hasn't been tampered by the time I'm finished doing a clean install
1879[22:59:48] <ryouma> debian was one of the first to ensure that packages are signed, and if you trust the install media, then you hould be able to trust the tools' provenance, provided no interference in the meantime
1881[23:00:01] *** Quits: Tempesta (Tempesta@replaced-ip) (Quit: See ya!)
1882[23:00:36] <ryouma> what would integrity checkers do that is more than they do? go to the web and use a different source? you have to assume the integrity checkers themselves are ok.
1883[23:00:54] <BuenGenio> exactly my point
1884[23:01:06] <ryouma> oh i thought you aewer asking a question rather than making a point
1885[23:01:16] <ryouma> never mind
1886[23:01:20] *** Quits: _till_ (~till@replaced-ip) (Remote host closed the connection)
1887[23:02:00] *** Quits: elibrokeit (eschwartz@replaced-ip) (Quit: A random quit message)
1891[23:02:51] *** Quits: bariscant (~bariscant@replaced-ip) (Remote host closed the connection)
1892[23:02:55] <BuenGenio> both :) the question part of that sentiment was - how do I automate the process of verifying the sums are ok once I'm satisfied I have an unadulterated connection to a mirror I trust
1893[23:03:45] <karlpinc> BuenGenio: G
1894[23:03:54] <BuenGenio> H-var, not after I've read this: replaced-url
1895[23:03:54] <BuenGenio> lol
1896[23:03:55] <karlpinc> That depends on what you mean by "automate".
1898[23:05:32] <karlpinc> BuenGenio: You could write a shell script that does a sha256 (or whatever the command might be to come up with the hash you desire) and prints the result and you manually eyeball it and see it checks. Or you could hire a firm in Banglore to make you a web page that will compare the checksum for you. How automated is automated?
1899[23:05:48] <BuenGenio> karlpinc, a one liner or a few statements to invoke a tool that would go through every binary/executable/resource file that's part of a package and cross check that it's identical to a package that hasn't been anywhere near my machine
1904[23:06:51] <karlpinc> BuenGenio: Usually you'd just check the checksum on the install media. But there's "debsums" that does what you want. (If I understand what you want.)
1905[23:07:28] <jhutchins> BuenGenio: We're assuming that you can find them easily.
1906[23:07:50] <BuenGenio> that's what I kind of want to know - but given I'm most likely not the first person who's after a similar goal, perhaps there's a tool I should and don't know about that may be already included as part of my stock distro
1907[23:08:10] <karlpinc> BuenGenio: That's the only problem with checksums. You have to assume that where ever you're getting them from is secure. So, get them from, say, 10 mirrors and the main debian website. (Using secure dns and https, or such.) And if they all match then you're pretty sure they're not compromised.
1911[23:09:09] <ryouma> but in any case debsums itself would be compromised
1912[23:09:33] <karlpinc> ryouma: But how complicated can debsums itself be? You could always read the source.
1913[23:09:39] <ryouma> the question is what dos one do to be sneaky (iu.e. something htat a bad guy cannot find automatically) while at the same time being a released package, i guess
1914[23:09:54] <ryouma> karlpinc: but then you have to trust the compiuler
1915[23:10:13] <ryouma> it's turtles alllllllllllllllllllll down
1916[23:10:27] <BuenGenio> karlpinc, debsums is really where I'm leaning towards, the only problem with debsums - or me - is that I don't see in the manual that I can instruct it to avoid checking against the sums in the debs that are on disk, and pull them from the mirror instead
1917[23:10:34] <karlpinc> ryouma: Interpreter. (perl?) So re-write it in another scripting language and double check.
1918[23:11:02] <karlpinc> BuenGenio: Hack the script.
1919[23:11:04] <ryouma> yes he could do that. bt he is asking for a turnkey soluytion.
1920[23:11:07] <BuenGenio> but if that were the ase that seems like a pretty trivial thing to "fix" with a script
1921[23:11:10] <BuenGenio> yeh
1922[23:11:38] <BuenGenio> karlpinc, good to be around people who are on the same wavelength :)
1940[23:15:59] <ryouma> or are you referring to the fact that you have to apt-get upgrade after the installation and some of the upgraded packages might be bad?
1962[23:28:50] <BuenGenio> @ryouma, well, other than the fact I know I had S3 boot scripts in my NVRAM, which I don't know how to get out and that I invariably end up having some if not most some sort of virtualisation/containerisation toolkit talking to random IPs in china and EC2 via encrypted UDP or mcast traffic or having a docker overlay of my whole FS mirrored to Amazon or Google Cloud, and snap that's fresh off the USB ISO burner laced with shit like this:
1970[23:32:45] <BuenGenio> like literally was the least succeptible or paranoid person I know until last december, when I realised I had an instance of ngrok running from inside one of the library folders I was using or a project I'd spent 6 months working flat out. started digging and discovered most of my servers were infested with cryptominers, loggers, database sniffers - you name it
1973[23:34:14] <BuenGenio> actually started a few weeks before that actually, when I realised I almost had to shout talking to my wife one evening because of how loud the fans in her Macbook were running
1983[23:38:26] <BuenGenio> literally tested out every freaking distro out there so far this year - Arch, Debian, Ubuntu, SuSe, Void, Guix and that super secure one from Redhat's, can't remember
1989[23:43:53] <BuenGenio> oh, did I mention that a couple of days after finally managing to put UB-ports on my phone... left it running for a couple of days, came back one day to see htop filled up with apache threads serving away gigs of encrypted traffic....
1993[23:46:55] <jhutchins> BuenGenio: While I'm sure you've found the process interesting, this really isn't the place to talk about all the OTHER Distros you've tried. If you have a debuan support question we'll be glad to help, bit fpr the rest of it there's #debian-offtopic.
1998[23:50:35] <BuenGenio> jhutchins, i obviously went off on quite a tangent, but has anyone else been noticing something not quite right - mostly on systemd based systems
1999[23:50:39] <BuenGenio> lately?
2000[23:51:12] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)