IRC log of #maemo-ssu for Saturday, 2013-06-15

freemangordonPali: I think we should make a bug triage meeting soon. I thinks most of the CVE bugs are fixed, but we need to check that00:01
freemangordonthink even :)00:01
Paliok00:01
_rdI ended in the reboot loop described here: http://talk.maemo.org/showthread.php?t=90385&page=4 after the last CSSU update.00:03
_rdI booted now with rescueOS and I read that having BatteryGraph installed is causing the reboot loop.00:04
_rdI now checked my /mnt/maemo/etc/init.d but I do not find a batterygraph startup script in there.00:04
_rdI am wonding if it is understood what is causing the problems and what I can do to fix it.00:05
kerioevent.d00:05
kerionot init.d00:05
kerioPali: *poke*00:06
Paliwhat?00:06
_rdthanks kerio.00:06
keriowhat _rd said00:06
_rdls bat* shows in ther battery-eye-daemon and batterypatch-check, no batterygraph.00:07
Palibootloop? remove SD card and update to last CSSU T8.200:07
Paliproblem with sd card was fixed in T8.100:07
_rdI had no SD card in the device00:08
_rdHow do I update, when I am in the bootloop?00:08
_rdchroot and then bring up networking manually?00:08
_rdchroot from rescueOS I mean00:08
Palihm, then I do not know... I know only about bootloop caused by inserted SD card00:10
_rdCan I find out what is causing the bootloop in my case?00:12
_rdWhat was causing the bootloop? Was it related to batterygraph as claimed in http://talk.maemo.org/showthread.php?t=90385&page=400:16
_rd?00:16
*** BCMM has joined #maemo-ssu00:17
Drathirok T8.2 on the board... update without problems for first look...00:30
*** nox- has joined #maemo-ssu00:30
Drathiri have maybe strange question from techniccaly side it will be hard to implement switch in security meno to not showing numbers only dot-es when type a security code?00:36
Drathirtechnically* menu*00:37
*** sixwheeledbeast has quit IRC00:43
PaliDrathir: that input widget is closed, so it is not possible to change00:43
Pali_rd: I do not know what batterygraph doing with bootloop00:44
*** dhbiker has quit IRC00:44
*** BCMM has quit IRC00:45
*** sixwheeledbeast has joined #maemo-ssu00:45
*** BCMM has joined #maemo-ssu00:46
_rdPali, is there a good way to determine what causes the bootloop?00:46
Palisyslog00:46
DocScrutinizer05disable watchdogs00:47
DrathirPali: oh little sad about that... because i also think about second feautures i mean button upon the done button with shutdown phone, harder is hit from mistake a button than all free place and phone goin to ask about shutdown...00:48
*** Mihanizat0r has quit IRC00:49
Drathiryea i know that only my imagines...00:50
_rdPali, what do I need to install for that? sysklogd?00:50
Paliyes00:50
_rdWould I copy the debian package on the n900 and then chroot into the maemo root dir and then install teh sysklogd package?00:51
DocScrutinizer05errr no00:51
DocScrutinizer05apt-get install sysklogd00:52
_rdbut my n900 does not boot anymore00:52
*** DaRoDi has quit IRC00:52
_rdOnly with rescueOS00:52
_rdor is there another way to boot up my n900?00:52
*** kolp has quit IRC00:52
DocScrutinizer05ooh I reread what you wrote00:52
DocScrutinizer05yeah might work00:53
DocScrutinizer05I somehow read chroot to easydeb00:53
*** tadzik has quit IRC00:54
*** tadzik has joined #maemo-ssu00:54
_rdah, ok. I will try that tomorrow. Many thanks.00:55
*** Pali has quit IRC00:56
*** _rd has quit IRC01:02
*** Vlad_on_the_road has quit IRC01:07
*** infobot has quit IRC01:51
*** infobot has joined #maemo-ssu01:52
*** ChanServ sets mode: +v infobot01:52
*** futpib has quit IRC01:59
*** LauRoman has quit IRC02:03
*** sixwheeledbeast has quit IRC02:08
*** NIN101 has quit IRC02:28
*** xes has quit IRC02:29
*** BCMM has quit IRC03:14
*** andre__ has quit IRC03:32
*** andre__ has joined #maemo-ssu03:33
*** andre__ has joined #maemo-ssu03:33
*** Martix_ has joined #maemo-ssu03:41
*** andre__ has quit IRC03:49
*** jon_y has joined #maemo-ssu04:05
*** nox- has quit IRC04:06
*** Martix_ has quit IRC05:28
*** amiconn_ has joined #maemo-ssu05:44
*** amiconn has quit IRC05:44
*** amiconn_ is now known as amiconn05:44
*** M13 has joined #maemo-ssu08:05
*** LauRoman has joined #maemo-ssu08:49
*** sixwheeledbeast has joined #maemo-ssu09:33
*** lartza_ has quit IRC09:54
*** lartza_ has joined #maemo-ssu10:01
*** lartza_ has quit IRC10:15
*** M13 has quit IRC10:15
*** LauRoman has quit IRC10:22
*** lartza_ has joined #maemo-ssu10:24
*** _rd has joined #maemo-ssu10:33
*** lartza_ has quit IRC10:37
*** lartza_ has joined #maemo-ssu10:44
*** _rd has quit IRC10:54
*** Vlad_on_the_road has joined #maemo-ssu10:55
*** Martix_ has joined #maemo-ssu11:00
*** kolp has joined #maemo-ssu11:24
*** fw190 has joined #maemo-ssu11:27
chem|stmorning...11:29
chem|stso lads and ladies...11:29
fw190morning11:29
chem|stI was thinking about... fuck freedom of choice... I will announce that properly, so noone can moan! if people want freedom of choice they need to code it themself, cssuS will get a minor upgrade with the pressing things and after that I will pull anything ~stable from testing11:31
freemangordonchem|st: hmm since when there is a freedom of choice in CSSU? :)11:32
*** futpib has joined #maemo-ssu11:32
chem|stfreemangordon: I do not pull some stuff from testing to give a choice, like the camera11:32
chem|stmerlin1991 is ok with it as I will get the shitstorm and me has to live with it!11:33
freemangordonaah, I see. well, fuck the freedom of choice then11:34
freemangordon:D11:34
chem|stwell I did not like it in the first place but with nokia upgrades you don't have that freedom so why should I provide it11:35
fw190chem|st: It wont be that  bad - eventually everyone will go for thumb and all the shtstorm will go on freemangordon :P11:35
freemangordonchem|st: that way I can start pushing for merging -thumb with -T11:35
chem|stfreemangordon: oh nope!11:35
freemangordonchem|st: exactly my point11:35
freemangordon(nokia)11:35
*** sixwheeledbeast has quit IRC11:36
freemangordonchem|st: hmm? I'll try to have KSSU in the next -T, I see no reason why not have thumbified packages slowly moving to -T after that11:36
chem|styeah slowly is the word I missed in the above11:37
freemangordonchem|st: -thumb is out in the wild for more than an year, benefits are clear, it is rock stable, etc11:37
freemangordonchem|st: comeon, have some trust in me :P11:37
chem|stfreemangordon: sry bu rock stable is without hardcore testers...11:38
chem|stI know it is stable11:38
freemangordonchem|st: no better test than wild11:38
freemangordonand it is in the wild11:38
freemangordonhowever11:39
freemangordonthat is for the future11:39
*** _rd has joined #maemo-ssu11:39
chem|stare you using your thumb device as daily phone?11:39
freemangordonofc11:39
chem|stok11:40
freemangordonI never issue an update to -thumb before havein the new packages installed on my daily device for a week or so11:40
freemangordon*having11:40
chem|stmy playtoy has testing and looses its battery without using it in a day11:40
chem|stidle with no simcard no wifi connected11:40
freemangordonthat's bad11:40
freemangordonbad battery?11:41
chem|stnope my sister has the very same, just not so frequent11:41
chem|stsomething hangs I guess11:41
chem|stI cannot find anything odd so I guess something is keeping some hardware powered without the system being aware of...11:42
chem|stsometimes it seems like the charging logic keeps itself powered after disconnecting11:43
chem|stor somelike11:43
freemangordonchem|st: that's weird, my devel device lasts a 4-5 days in the drawer, with wifi on11:44
chem|stwell I have to admit it is my playtoy and as soon as I roll over a bcakup it might be gone^^11:44
chem|stbut I didn't alter anything but turning off things... I could dismantle it and check if the screen stays powered11:45
freemangordonchem|st: do you have rd mode on?11:46
chem|stnope11:46
freemangordonhmm. checked with powertop?11:46
chem|stsure11:46
chem|stas I said it is nothing obvious11:46
freemangordondid you try with different battery?11:47
chem|stit might be some device that I turned off by hand... if the kernel does not control it anymore it might get fully powered11:47
freemangordonit might worth thr try11:47
chem|sthad that with my wifi on my laptop11:47
*** Martix_ has quit IRC11:51
*** LauRoman has joined #maemo-ssu11:52
merlin1991heh I didn't say I'm ok with it, I said it isn't my call anymore :D11:55
freemangordonoh you...11:56
*** _rd has quit IRC11:58
*** _rd has joined #maemo-ssu11:58
*** _rd has quit IRC12:02
*** M13 has joined #maemo-ssu12:03
*** sixwheeledbeast has joined #maemo-ssu12:05
*** NIN101 has joined #maemo-ssu12:10
sixwheeledbeastSo should CSSU-S go thumb only in the future? I don't see why not myself.12:10
keriowell, it makes it really awkward for the user12:12
keriobecause at that point you have to be careful about the kernel12:13
*** _rd has joined #maemo-ssu12:17
sixwheeledbeastkerio: but KCSSU will be a dependency?12:23
sixwheeledbeastIf there was an option of CSSU-S and CSSU-S+thumb. You would go with the benefits of thumb.12:29
sixwheeledbeastsurely?12:29
sixwheeledbeastSo it seems silly to work on both IMO12:30
*** BCMM has joined #maemo-ssu12:39
*** fw190 has quit IRC12:48
*** futpib has quit IRC13:00
*** sixwheeledbeast has quit IRC13:10
*** LauRoman has quit IRC13:13
*** jon_y has quit IRC13:33
*** jon_y has joined #maemo-ssu13:34
*** sixwheeledbeast has joined #maemo-ssu13:35
*** sixwheeledbeast has quit IRC13:51
*** sixwheeledbeast has joined #maemo-ssu14:03
Drathirchem|st: on power saving turn less bright of lcd, and auto device lock14:12
Drathirmine with using and wifi works few hours, but not used can be on 1day14:14
*** sixwheeledbeast has quit IRC14:16
*** sixwheeledbeast has joined #maemo-ssu14:26
*** Milhouse has quit IRC14:32
*** Milhouse has joined #maemo-ssu14:47
*** lartza_ has quit IRC14:50
*** lartza_ has joined #maemo-ssu14:58
*** unclouded has quit IRC15:02
*** NIN101 has quit IRC15:05
*** NIN102 has joined #maemo-ssu15:06
*** _rd has quit IRC15:18
*** _rd has joined #maemo-ssu15:19
chem|stDrathir: device is idle, screen off, flightmode... and still15:21
Drathirhmmm thats interesting... if on clean device is the same maybe is a hw issue?15:31
Drathiryou also can check cpu usage usage with widget...15:33
*** M13 has quit IRC15:36
*** M13 has joined #maemo-ssu15:37
*** futpib has joined #maemo-ssu15:38
*** _rd has quit IRC15:39
*** _rd has joined #maemo-ssu16:09
DocScrutinizer05sixwheeledbeast: IF thumb is such low effort and as compatible, then it's just about compiling everything with -thumb flag in builder set16:17
DocScrutinizer05and then it's not silly to work on both but actually sensible to offer that alternative16:17
DocScrutinizer05but requiring all cssu-s users to replace their kernel is sth we already discussed and found will give us problems16:18
sixwheeledbeastproblems how?16:20
keriofor instance, you need to force people to replace their kernel16:23
sixwheeledbeastAnd that's an issue because?16:24
DocScrutinizer05because they might not even be able to do so16:25
DocScrutinizer05where's the uBoot with thumb kernel?16:25
DocScrutinizer05where are the kernel.bz images that are thumb enabled?16:25
DocScrutinizer05basically changing the thumb thingie starts a complete new distro,16:26
DocScrutinizer05while CSSU is still an *update* to stock distro16:26
DocScrutinizer05going thumb creates a system that's fundamentally incompatible to stock maemo16:27
sixwheeledbeastah16:27
DocScrutinizer05installing uBoot on your thumb system will fubar it16:27
DocScrutinizer05installing multiboot on your thumb system will probably fubar it16:28
DocScrutinizer05you dunno what else depends on stock kernel, or is based on stock kernel16:28
sixwheeledbeastI don't use multiboot :)16:28
DocScrutinizer05that's irrelevant16:29
DocScrutinizer05brainfuck-kernel is also incompatible to thumb I guess16:29
DocScrutinizer05honestly unless you roll out a thumb (and nothing else but thumb) kernel via Nokia stock repo, as PR1.4, so *all* devices do the update, I don't see how CSSU which is meant to be the bugfix addon and continuation of what's in stock repo can ever go thumb per default16:32
sixwheeledbeastbut hang on KP has boot variants why not KCSSU16:32
DocScrutinizer05hmmm?16:32
DocScrutinizer05what's the use of KCSSU boot variants when you installed brainfuck-scheduler-kernel which doesn't know shit about thimb?16:33
kerioDocScrutinizer05: fwiw, we can just upgrade everything together16:34
kerioand remove the braindead scheduler16:34
DocScrutinizer05yes, that's what you call a new distro16:34
kerionaaaaah16:34
keriothe only problem is that we don't yet control the nokia repos16:34
*** futpib has quit IRC16:35
*** sixwheeledbeast has quit IRC16:37
DocScrutinizer05yes, when we control Nokia repos, we can actually roll a PR1.4, and show uBoot and multiboot and brainfuck the finger, since then it's them to either stay on PR1.3 or patch their software to upgrade to PR1.416:38
jon_ywhats wrong with uboot?16:38
jon_yhow are you supposed to boot off SD?16:38
DocScrutinizer05then every user out there is free to accept the system upgrade and know he can install whatever he likes form whatever repo and it will work, or he rejects the PR1.4 system upgrade and nails his system to PR1.3 and then the "old" kernel dependent stuff from PR1.3 repos will still work for him16:39
DocScrutinizer05eh?16:40
*** _rd has quit IRC16:40
DocScrutinizer05what's wrong with uBoot? it's simply a bloated mess that tries to be a OS of its own. I dunno what else is wrong with it16:40
*** _rd has joined #maemo-ssu16:41
DocScrutinizer05kerio: it's as plain simple as that: such a non-backward-compatible change in core sysem components creates a new OS major revision, like frooblux23.4 -> frooblux24.016:44
keriopr 1.4 then! :D16:45
DocScrutinizer05that's exactly what happens also on Maemo PR1.2 -> PR1.316:45
DocScrutinizer05and yes, we need a PR1.4 then to enable thumb16:45
DocScrutinizer05you are aware that _this_ Pr1.4 would obsolete CSSU-S?16:48
DocScrutinizer05since it IS CSSU-S16:48
keriocssu-1.416:49
DocScrutinizer05>>It aims to deliver fixes which would be difficult to deliver through Extras (like core Maemo packages).<< It's not difficult at all to deliver those packages thru "Nokia" repos16:49
DocScrutinizer05once we gain full control over those repos16:50
DocScrutinizer05but I still don't see that happen16:50
jon_yDocScrutinizer05: I don't see other alternatives to uboot16:52
DocScrutinizer05unless we nuke the original Nokia repos from user's apt list and replace them with cssu-core repos, on installation of CSSU-1.416:52
DocScrutinizer05jon_y: me neither16:52
DocScrutinizer05I neither see anything wrong with uBoot16:52
jon_y"it's simply a bloated mess that tries to be a OS of its own" << ?16:53
jon_yisn't it supposed to work that way? a bootloader?16:53
DocScrutinizer05you asked "what's wrong with uBoot?" so I answered16:53
jon_y"and show uBoot and multiboot and brainfuck the finger" << I assumed you wanted to nuke uboot?16:53
DocScrutinizer05WUT? nooo you got that wrong16:54
DocScrutinizer05it's just that with an "official" PR1.4 it's the maintainers of those packages that have a problem to update their software to new kernel, not we (the cssu maintainers)16:55
jon_yok16:55
DocScrutinizer05it's just that PR1.3 depends on kernel "2.6.28-omap1 #1" and the uBoot and other packages depend on PR1.3 for the kernel compatibility16:58
kerioi'm not entirely sure why uboot ships with an embedded kernel, actually16:58
DocScrutinizer05because it replaces original kernel16:58
kerioso? install a bootimg16:58
DocScrutinizer05so it needs *some* kernel to ship with it, to boot the system after installing uBoot16:59
DocScrutinizer05kerio: are you already awake?17:00
kerioi am17:00
DocScrutinizer05kerio: where's that bootimg in FIASCO?17:00
keriowhy fiasco?17:00
kerioit's not preinstalled, no17:00
keriobut neither is uboot17:00
DocScrutinizer05no, you're not ;-P17:00
kerioyou definetely need some kernel17:01
kerioi'm challenging "you need an attached kernel"17:01
DocScrutinizer05you're changing default system architecture17:01
DocScrutinizer05that's almost worse than the switch to thumb17:02
DocScrutinizer05;-P17:02
kerioyou're just changing bootloaders17:02
keriothe new bootloader needs the kernel in a different place17:02
kerioit's like moving from lilo to grub17:02
DocScrutinizer05well, yes17:02
DocScrutinizer05and that neither is sth done lightly17:02
*** sixwheeledbeast has joined #maemo-ssu17:03
DocScrutinizer05again, where is your kernelimg in flasher-3.5 -rootfs-only -Ff anyrandomCOMBINED.bin ?17:04
*** _rd has quit IRC17:04
*** _rd has joined #maemo-ssu17:04
kerioyou never have to flash the kernel again, actually17:04
kerioyou just put it on a uSD and boot from the console, in case of emergency17:04
DocScrutinizer05sure you have to, with *every* flashing you do17:05
DocScrutinizer05since... see above, just 2 lines above17:05
DocScrutinizer05and you'd be surprised how many devices with broken uSD slot are out there17:06
DocScrutinizer05and actually now you're not only not awake, you're actually having a bad dream. Make my device boot-fail when I swap uSD, are you NUTS??17:06
keriono, that's just in case of emergency17:07
keriothe kernel is on emmc17:07
DocScrutinizer05think about that again please!17:07
_rdI am getting close to get a syslog from my n900 in the bootloop since the last CSSU....17:08
*** sixwheeledbeast has quit IRC17:08
DocScrutinizer05what we actually could (and I already did) consider: placing kernel to initrd NAND partition17:08
DocScrutinizer05mtd4 iirc17:08
DocScrutinizer05afaik definitely nothing is using initrd mtd4 on N90017:12
DocScrutinizer05and it#s not swappable ;-P17:12
DocScrutinizer05unlike uSD17:13
DocScrutinizer05and shouldn't usually get nuked on flashing a new rootfs17:13
kerioi just don't get why you dislike my setup17:18
keriothe kernel isn't flashed, it's just written to the eMMC17:18
kerioi mean, it's a file in MyDocs17:19
_rdHmm, it seems my X server is not coming up since my last CSSU update: http://bokomoko.de/~rd/syslog17:22
_rdAs a side note: it is possible to install syslog with rescueOS....but painful.17:22
_rdDo you have any idea why that happens?17:23
kerioJan  1 07:38:32 Nokia-N900 init: xomap post-start process (922) terminated with status 12717:24
kerioJan  1 07:38:32 Nokia-N900 DSME: Process '/usr/bin/Xorg -logfile /tmp/Xorg.0.log -logverbose 1 -nolisten tcp -noreset -s 0 -core' with pid 919 exited with r17:24
_rdjust wondering if I could get the Xorg.0.log17:28
_rddid anything change related to X in CSSU?17:33
*** futpib has joined #maemo-ssu17:40
*** futpib has quit IRC18:37
DocScrutinizer05kernel in Mydocs??19:28
DocScrutinizer05that sounds incredibly great, particularly to umount MyDocs for exporting it via ass rage mode, and eventually completely cleaning the volume19:29
*** iDont has joined #maemo-ssu19:31
DocScrutinizer05_rd: shouldn't19:34
DocScrutinizer05_rd: replace your transitions.ini by the backup19:35
DocScrutinizer05remove your SIM19:35
DocScrutinizer05and maybe even mv /home/user /home/user-AWAY; mkdir /home/user19:37
DocScrutinizer05also df -h /19:38
DocScrutinizer05filled rootfs is the most usual reason for effects like yours19:38
*** sixwheeledbeast has joined #maemo-ssu19:39
_rdDocScrutinizer05, rootfs looks good:19:56
_rdrootfs                  233344    179628     49624  78% /19:56
_rdnone of the fs is more filled than 90%19:57
*** iDont has quit IRC20:01
_rdtransitions.ini is in /usr/share/hildon-desktop, right?20:01
sixwheeledbeast~transitions.ini20:02
infobottransitions.ini is, like, http://wiki.maemo.org/Community_SSU/Features/hildon-desktop20:02
sixwheeledbeastyep20:03
_rdI noticed that I have a transitions.ini.dpkg-dist from May 24th, 2013, transistions.ini is from 201120:07
_rdI wanted to rename /tmp/Xorg.0.log to /Xorg.0.log or /home/user/Xorg.0.log, that it does not get deleted at reboot, but that for some reason did not work. Does anybody know why?20:07
DocScrutinizer05rename? how?20:08
_rdin /etc/event.d/xomap .... I wanted to see what is in the log file20:09
_rdchecking new transitions.ini now20:09
DocScrutinizer05if that's the line that uses dsmetool to start Xorg then it's the right location20:09
DocScrutinizer05when your transition.ini is old, then it probably didn't change and thus not the culprit20:10
_rdyes it is the dsmetool line, but that for some reason did not work.20:10
DocScrutinizer05how "doesn't work"?20:11
_rdI did not see the reboot cycles anymore, but it also did not come up.20:11
_rdand I never found an /Xorg.0.log20:11
DocScrutinizer05editing that line to point to /home/user/Xorg.0.log should work20:12
DocScrutinizer05probably Xorg runs as user, and user mustn't write to /20:12
_rdI tried that as well, but I had also no luck with that.20:12
DocScrutinizer05and MyDocs might fail for other reasons20:12
_rdyou recommend to boot without SIM, right?20:13
DocScrutinizer05or you create a dir /mylogs; chmod a+rw /mylohgs; then point the line to there20:13
DocScrutinizer05right20:14
_rdthat is a good idea20:14
_rddoes an unavailable SIM delay the boot cycle?20:14
DocScrutinizer05it's been seen occasionally that a rogue or crippled SMS on SIM made the whole system go bonkers20:14
DocScrutinizer05shouldn't delay20:14
_rdI do not see the reboots anymore, but also no boot progress.20:15
_rdlet me boot in rescueOS again and create the /mylogs20:15
*** rd_ has joined #maemo-ssu20:20
*** _rd has quit IRC20:22
rd_Xorg.0.log is hidding....20:27
rd_....with /mylogs I have the same behaviour as with /tmp but no Xorg.0.log is in there.20:28
rd_can anybody check if on a regularly running device  /tmp/Xorg.0.log exists?20:29
DocScrutinizer05IroN900:/home/user# ll /tmp/Xorg*20:38
DocScrutinizer05-rw-r--r-- 1 root root 4521 2013-06-13 15:09 /tmp/Xorg.0.log20:38
DocScrutinizer05just for fun try Xorg logging to /dev/null20:39
rd_The only scenario which makes sense for me is that the xomap crashes before it can start writting a log20:39
rd_ok, I can do20:39
DocScrutinizer05so you can tell apart if it doesn't log because it crashes, or it crashed because it can't log20:40
*** sixwheeledbeast has quit IRC20:42
rd_just for curiosity: does "exec cat /mylogs/Xorg.0.log >> /mylogs/Xorg.log" in the pre-start script section hurt?20:43
rd_if I add this line, I do not get reboot loops anymore20:43
rd_....but the boot process seems also to hang.20:43
DocScrutinizer05haha20:43
DocScrutinizer05exec for sure is ... questionable here20:44
DocScrutinizer05you're aware exec aborts the script, like exit20:44
rd_ok, I tried before without...but no change20:44
rd_trying with /dev/null now20:45
rd_thanks for the explantion of exec20:45
DocScrutinizer05afk bbl20:45
rd_I also get reboot loops when using /dev/null20:46
DocScrutinizer05exec *replaces* the current process with the new one20:46
DocScrutinizer05(actually not the process, but that doesn't matter here)20:47
DocScrutinizer05afk bbl20:47
rd_and I assume it does not return there when it ends20:47
rd_strange...now the n900 only rebooted once, now it seems to hang again.20:48
rd_trying again20:49
*** rd_ has quit IRC21:05
*** rd_ has joined #maemo-ssu21:05
rd_I start to run out of ideas. Maybe I should just save the remaining data away and re-flash.... and hope that the problem goes away after re-flashing, but it came with the reboot of the CSSU update, so it is at least likely.21:09
*** arcean has joined #maemo-ssu21:21
DocScrutinizer05you most probably are missing files in ~user/.* now21:36
DocScrutinizer05you could try21:36
DocScrutinizer05echo "before Xorg" >>/mylogs/mylogs.txt21:37
DocScrutinizer05before dmsetool Xorg21:37
DocScrutinizer05and21:37
DocScrutinizer05echo "after Xorg" >>/mylogs/mylogs.txt21:37
DocScrutinizer05below it21:37
DocScrutinizer05what you consider "hanging" might actually be fsck trying to do its thing21:38
DocScrutinizer05which can take quiiiite a while21:39
kerioand it'll probably fail at it21:42
*** BCMM has quit IRC21:42
*** BCMM has joined #maemo-ssu21:42
rd_if it would be fsck, I would expect entries in /var/log/fsck_home.log21:53
rd_but they end all with clean21:53
DocScrutinizer05hmm21:53
rd_what is the point of after?21:53
rd_"after Xorg..."21:53
rd_you said that exec will replace the current process (?)21:54
DocScrutinizer05well, just in case21:54
rd_ok21:54
DocScrutinizer05after all dsmetool shouldn't block21:54
rd_Hmm...would they be regenerated if I move /home/user away?21:55
DocScrutinizer05should, afaik22:00
DocScrutinizer05wait, is that "exec dsmetool -foo Xorg... "?22:01
DocScrutinizer05indeed22:03
rd_Is that explaining some of the "miracles"?22:04
rd_moving /home/user away did not change anything.22:05
DocScrutinizer05nope22:06
DocScrutinizer05it's just making me wonder if i finally should learn a bit about upstart scripts and how they work22:07
DocScrutinizer05well, it's pretty obvious that dsme finds Xorg aborting for -n=8 times and then -r-eboots the device22:07
DocScrutinizer05so the problem is in Xorg itself22:08
DocScrutinizer05you tried placing the logging elsewhere (/dev/null) and that didn't help22:08
DocScrutinizer05so it's most probably not the logging that fails22:08
DocScrutinizer05since it seems Xorg doesn't log *anything* nevertheless, i'd guess sth elementary is broken22:09
DocScrutinizer05if the last test I suggested shows that you actually _can_ echo something >>/mylogs/foo.bar22:11
DocScrutinizer05then we also can rule out filesystem mounted r/o or whatever22:12
rd_yes, that works22:12
rd_I see an entry from before.22:12
rd_filesystem is definitely ok22:12
DocScrutinizer05then I'm also out of ideas22:12
DocScrutinizer05particularly since it seems Xorg doesn't log a single char22:13
rd_7f47cfbc6c9d0c87853be2229268fd9a  /usr/bin/Xorg22:13
rd_does that md5sum match yours22:14
rd_-rwxr-xr-x    1 root     root      1568060 Feb 16  2010 /usr/bin/Xorg22:14
DocScrutinizer05af3220748175c52b2fea9f3045b68d08  /usr/bin/Xorg22:14
DocScrutinizer05-rwxr-xr-x 1 root root 1525792 2010-02-16 14:15 /usr/bin/Xorg22:15
*** futpib has joined #maemo-ssu22:16
rd_Hmm that is weired?22:18
DocScrutinizer05yeah, a bit22:18
rd_what is your version of xserver-xorg-core 2:1.6.99.1-0osso20090208.108+0m5 ?22:18
rd_This contains Xorg22:18
DocScrutinizer05xserver-xorg-core:22:19
DocScrutinizer05  Installed: 2:1.6.99.1-0osso20090208.108+0m522:19
DocScrutinizer05virus? ;-P22:20
DocScrutinizer05I'd rather suspect you might have installed the thumb binary22:21
DocScrutinizer05wild guessing22:21
rd_Hmm...not that I am aware of....22:21
rd_trying a "apt-get install --reinstall xserver-xorg-core"22:22
rd_funny that the data is the same.22:22
rd_date22:22
rd_virus on the n900 would be really surprising.22:22
DocScrutinizer05indeed22:26
DocScrutinizer05hardly believable22:26
DocScrutinizer05and yep, date is quite strange22:26
rd_still reboot loops, but let's see if the Xorg.0.log is now created22:34
DocScrutinizer05I'd rather be interested in size of your Xorg now22:39
DocScrutinizer05quite possble that mine is a bit strange, not yours22:40
rd_no, after the reinstall I got your md5sum22:46
rd_....but I saved the old away for inspection ;-)22:46
rd_reinstallation also reinstalled /etc/event.d/xomap22:49
rd_The LED becomes sometimes green, I do not remember that I saw that before the reboot loops. Do you know what that means?22:51
DocScrutinizer05it means that bme and hald-addon-bme and mce and a few other systems decided that your device is on USB power and battery is completely charged22:53
DocScrutinizer05actually that would be interesting (if only for academic purposes, though you never know) to see if your device also goes into bootloop when plugged to charger when powered down22:55
rd_Hmm...the green LED comes when it is not on a charger, but it is not blinking either, just flashes once.22:57
rd_Also after some reboot loops the n900 seems to hang.22:57
DocScrutinizer05yeah, that's rather normal I think22:58
DocScrutinizer05the green flashing22:58
rd_ah, ok.22:58
rd_actually it is not hanging, it powers off22:59
Drathirfor mine obserwation blink grees is short between security and code pin but if battery have low power in this moment also can shutdown and boot again if im not wrong....23:00
DocScrutinizer05wait, what? you deleted /home?23:00
DocScrutinizer05ls -l /opt errr23:01
DocScrutinizer05you can't23:01
rd_shortly before the xomap crashes I see23:02
rd_Jan  1 01:16:25 Nokia-N900 ohmd[806]: GLIB ERROR ** default - module videoep failed to load but listed in require aborting...23:02
rd_Did CSSU changes in glib?23:02
DocScrutinizer05when you deleted /home, you're out of luck since that also nuked /home/opt/*23:02
DocScrutinizer05no23:02
DocScrutinizer05prolly not23:03
rd_no I did not delete home, I temporaryly move home/user away, but I moved it back23:03
DocScrutinizer05actually this isn't a cssu topic at all23:03
DocScrutinizer05or rather, most likely not23:03
rd_I agree, although it is strange, that I did the CSSU upgrade and the reboot required for the update ended in boot loops23:04
rd_BTW, do you also see the GLIB ERROR in your syslog?23:05
DocScrutinizer05yeah, that *might* be related to cssu T8 which had a fuxored fstab/fsck feature23:05
DocScrutinizer05that might have resulted in rootfs getting messed up, or whatever23:06
DocScrutinizer05but now the problem is not related to cssu anymore, it at best been caused by cssu23:06
DocScrutinizer05the problem not seems more generic23:06
DocScrutinizer05now*23:06
rd_How can I find out which CSSU I have installed?23:07
Drathirrd_: can i ask where you move home/user ?23:07
DocScrutinizer05hmm23:07
rd_mv /home/user /home/user.bak23:07
rd_mkdir /home/user23:07
Drathirok23:07
rd_chown user:users /home/user23:07
rd_<tested>23:07
rd_rm -rf /home/user23:07
rd_mv /home/user.bak /home/user23:08
Drathirim not sure if user isnt ext partition if moved to fat loosing permission when back but in this case ok...23:08
DocScrutinizer05nobody moved sth to vfat23:09
*** Martix_ has joined #maemo-ssu23:14
* DocScrutinizer05 ponders starting another BM backup23:18
DocScrutinizer05actually osso-backuo's inability to do automated backups is one of the most braindead things in maemo23:19
rd_Do you save away entire disk images, such that in case of an emergency you can play back the entire installation?23:20
rd_(or do a diff ;-)23:20
DocScrutinizer05BM does logical full images23:21
DocScrutinizer05not disk images23:21
DocScrutinizer05actually disk images are problemantic at least on NAND/ubifs23:21
*** LauRoman has joined #maemo-ssu23:22
rd_ok23:23
Drathirim not sure but backup menu could work in this case because it is before green light flash... or im wrong?23:23
DocScrutinizer05which is why BM-rev1 got discontinued23:23
rd_so thanks to rescueOS I at least did not loose data.23:23
rd_I will copy away everything overnight.23:24
DocScrutinizer05backup of MyDoc shouldn't take longer than 3h or so23:25
DocScrutinizer05iirc23:25
rd_yes, but in 3h I am sleeping :-)23:27
*** freemangordon has quit IRC23:43
*** M13 has quit IRC23:59

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!