You are not logged in.

#1 Re: Creating/Maintaining Packages » noreply@archlinux32.org sending Spam ? » 2019-02-18 20:47:49

Okay, I've seen the full email of ula8000 now.  It suggests google actually received the mail from an ipv6 address indicating a cloud computer run by a run by centrilogic, rather than tyzoid's server, which is different to andreas's mail above.

Andreas' mail looks to me all like a geniune mailing from the boards, the only odd thing is I can't find a user called Josephchich.  I'm not sure if I can see banned users in the user list though.

#2 Pacman / Pacman Upgrades » pandoc is broken in community-testing (haskell-quickcheck too new) » 2019-02-18 19:02:13

levi
Replies: 0

In the testing repos, haskell-quickcheck is at version 2.12.6, but pandoc is at version 2.5.

Pandoc doesn't actually have a dependency on haskell-quickcheck, but it tries to load a .so from it when it launches.  With the current testing configuration, this fails, because it's looking for an older version.

I had a copy of haskell-quickcheck version 2.11.3 in my package archive, so I downgraded to that and it fixed it.  I note there's a positively new version of pandoc currently in staging, and I don't know if that would properly use haskell-quickcheck 2.12.6 or not, but presently at least, the packages supplied in testing don't work.

#3 Re: Creating/Maintaining Packages » noreply@archlinux32.org sending Spam ? » 2019-02-18 17:11:49

Received: from mail.tyzoid.com (mail.tyzoid.com [23.92.211.62])
	by smtp.andreasbaumann.cc (Postfix) with ESMTP id 51234CD9FB
	for <mail@andreasbaumann.cc>; Mon, 18 Feb 2019 11:23:42 +0100 (CET)

This line at least claims to have been written by smtp.andreadbaumann.cc which I assume is under your indirect control at least.  That suggests it really was on a server called mail.tyzoid.com at the step before, so he's gone to a lot of effort faking it if that is all bogus.

#4 Re: Creating/Maintaining Packages » noreply@archlinux32.org sending Spam ? » 2019-02-17 17:00:01

Can you post the intitial Received: lines?  You shouldn't need to post any received lines from after it gets inside google's domain, but there may be some previous lines that indicate it's actual mail path  (although it's always possible to prepend some fake ones, they should be true after a point).  Decoding received lines is a bit of a black art I find, but gmail when I used to use it at least seemed to add headers like this:

Received-SPF: pass (google.com: best guess record for domain of #mailsrv# designates #ip# as permitted sender) client-ip=;
Authentication-Results: mx.google.com; spf=pass (google.com: best guess record for domain of #mailsrv# designates #ip# as permitted sender) smtp.mailfrom=#mailsrv#

If they're still adding those headers (my last email to check is from 2018 before google blocked my client), at least one of them should be true and might tell you the actual server used, or at least google's best guess.

#5 Re: Creating/Maintaining Packages » noreply@archlinux32.org sending Spam ? » 2019-02-16 19:53:27

I wonder about gmail adding the 'yes this is you' helper to your to field.  That suggests to me they didn't have your exact email address, but gmail guessed they meant you.  Can you check to see if the email address they used is exactly your address?

I've not had any spam like that yet here.  Maybe the database was leaked before I registered?

#6 Re: Creating/Maintaining Packages » extra/firefox » 2019-02-15 17:19:31

Yep, I've had no unexpected problems using it over the last two days.  The option to disable blocking per site isn't actually that useful to me, because I never want to run trackers, but I sometimes have to enable cookies, so I still have to do that by right clicking and hitting
'page info' and then modifying cookie settings for that page.  It's not got worse, so I can't really complain.  At least it's not crashed for me yet.

And thanks again for your efforts to actually port this over.

#7 Re: Kernel & Hardware » How to install linux-ck without compiling? » 2019-02-15 05:11:26

In my experince, allocate a few hours to compile a kernel.  It depends largely on the hardware you're running it on, of course, and the danger when talking about i686 binaries you're using a seriously old computer, but it would have to be really pretty ancient to take a whole day.  I suspect most things built using make are io-limited most of the time anyway, so the more important thing is probably how fast your discs are.

#8 Re: Pacman / Pacman Upgrades » [solved] 11.02.2019 Error "pacman -Syu" » 2019-02-12 20:47:11

It may be worth doing a second -y when you've changed your mirrorlist file, but I personally have never found the need.

I guess as long as you pacman -Syu appears to be downloading new lists periodically it's all okay.  My own system seems to think everything's been up to date for the past month, so I ran my mirrorlist.pacnew through rankmirrors and updated my mirrorlist.  That got me a whole raft of updated packages, so I assume the old server that happened to be on top of my mirrorlist had stopped updating.

I still didn't need to do a -yy to get new package lists.

I would guess that you using -yy has not had anything to do with your improved experience, just that whatever problem you had has happened to have been fixed since your last -Syu.

#9 Re: Pacman / Pacman Upgrades » exfat-utils 1.3.0-1.0 won't mount drives with udevil » 2019-02-12 20:37:04

As far as I can tell, neither udevil or exfat-utils have been updated since before last November.  Is this a new bug, or have you just started using this?

#10 Re: Creating/Maintaining Packages » extra/firefox » 2019-02-12 17:40:11

Thanks for your work on this.  Is there any hope that the binaries you manage to produce are less crashy than the ones mozilla produce?

Edit: I'm posting this from firefox 65.0 that happened to come down via a -Syu.  I don't really understand what's gone on to enable this, but I'm glad to be able to use the new tracking protections.

#11 Re: Creating/Maintaining Packages » extra/firefox » 2019-02-10 19:20:26

FWIW, I just tried installing the 65.0 binaries from mozilla using the non-Tim Song pkgbuild.  It seemed to have more shasums than I could really cope with (four different ones) so I just skipped them, but on successfully installing it and running it for a few hours, it just sat there allocating memory but never actually using it all the while spinning up one half of my processor (one hyperthreaded core) at pretty much 100%.  No logs even with verbose, and nothing in my logs either.  I'm not in the mood to debug it further, so I've switched back to 63.0 to post this.

#12 Re: Mirrors » Online Mirrorlist » 2019-02-10 15:56:04

I'll just note here that the search form on the packages page hasn't survived the migration intact still.  I think it's this line that's at fault:

<form id="pkg-search" method="get" action="/">

(from within div id="pkglist-search")

Action should be "/packages/", as far as I can remember these things. Certainly, when doing a search and it dropping you to the wrong page, inserting /packages after the hostname is a valid workaround.

#13 Re: Mirrors » Online Mirrorlist » 2019-02-08 16:04:41

Yeah, I think there's a bit of movement on the packages index, and perhaps the mirrors too; I posted above a couple of days ago about bad certs, but since then it's moved from packackes.archlinux32.org to archlinux32.org/packages which fixed that, but has broken the search button which still submits to root, which is no longer valid.

But I didn't complain yet, because it was still a moving target. If I'd spotted it at the weekend I might have put it in IRC when I don't mind going on there, but moaning here had a bit too much permanence for something so transitory (hopefully).

On the topic of iso mirrors versus package mirrors, I think for the time being they're essentially identical sets.  They're basically just a little server space, and if you've the space to host all of the packages you're going to have the space to store a few isos.  And the other way round, I don't think anyone cares about the project but not enough to just help people get installed but not support them ongoing.

Edit:

I've never used reflector personally.  I just get a mirrorlist with my pacman updates and then consider editing it and using rankmirrors on the remainder if there's a lot still.  But it is supplied with arch32 (just, after doing a search and getting to the wrong page, put 'packages' in the url after the hostname and slash, like I have there), so it not working is a bona fide bug and should probably be raised in the bug tracker if you've got set up there already.

#14 Re: Creating/Maintaining Packages » extra/firefox » 2019-02-07 16:51:11

deep42thought wrote:

there are builds scheduled and failing
I didn't have a look into them yet, sry

Well, thanks in advance for when you do.

FWIW, today I'm getting an HSTS key error because my browser claims the cert's only valid for archlinux32.org and www.the same when I try that link or go to packages.... manually.  It seemed to be working when I used it yesterday, so I'm not sure what's changed here.

I'll have to try out the prebuilt firefoxes on my machine and see if I can reproduce the reported errors

#15 Re: Creating/Maintaining Packages » extra/firefox » 2019-02-07 04:19:47

And as of a week ago last Tuesday we're even further behind current, with firefox 65.0 being released.  I can't see a build error for firefox, so I guess it's just not building automatically at the moment.  I may have to switch back to using one of the aur projects that pulls down mozilla's builds.

#16 Re: Mirrors » Online Mirrorlist » 2019-02-04 22:17:12

When I posted in this thread at 2019-01-22 21:04:39, I checked that simplysam.us link and it was serving old kernels.  Now I check it it's up to date, but it was causing a user problems before now.

#17 Re: Installation » Advice for partitioning a new 1 TB hard disk » 2019-02-02 21:44:52

FWIW, I've just upgraded my server box (running arch64, but I'm sure the space would be plenty for arch32 bins too) using a TOSHIBA DT01ACA100 (snap) that I had been using for backups, but since then I bought a 3TB drive when the price per gb for that size became the lowest.  I just copied over my the contents of my old partitions to the new drive, reinstalled grub, fiddled with the fstab and hostname files (gave it a different hostname, just for fun).  There's some helpful tips on that Installing arch from another linux article that I'd missed, primarily using the -x option to cp to stop it copying over recursive mountpoints, which would have simplified things for me and maybe stopped me accidentally copying my /boot folder contents to /boot/boot, which cause a fair bit of headscratching*, but I've finally fixed that after a few days of failing to spot my mistake.

My partitions were 16GB for root, 8GB for /var 64MB for /boot and the rest for /home on my older 320GB drive.  For this update I've doubled boot to 128MB because it was looking a little tight, and updated my /var to 16GB (I've not deleted my pacman cache since early 2016 apparently, and in my experience your /var/cache/pacman/pkg folder is the vast majority of any /var folder).

Just put the final bolt in and secured the tin box that is my server (actually an old desktop chassis, meaning I had to take out half of the cabling on the motherboard before I could manoeuvrer the old hard disc out).  It's emitting a bit of a hum at the moment (unsurprising when it's spinning at 7200rpm, I guess) so I'll have to see if I can live with that.

* When you've not got a kernel in the right place in /boot, there's no symptom when you chroot onto it because it's still running the parent kernel, all that really happens is grub-mkconfig just prints out a config file with no menu items in it.  I copied my old grub.cfg over and tweaked the uuid and got it to boot so far, but when it told be it couldn't find the kernel, it took me a surprisingly long time to actually check it out.

#18 Re: System Administration » Ruby 2.6 Rubygems error » 2019-02-01 21:07:15

Can you check your 2.6 and see if it includes rubygems.rb somewhere.  Something like 'bdstar -tzf /var/cache/pacman/pkg/ruby-2.6*|grep rubygems.rb'

And see if you've got rubygems.rb in 2.5, or if this is a newly required inclusion.  You could do something similar for that, or look in your filesystems, but on my installs at least, filtering through everything in /usr is generally slower than listing a package.

#19 Re: System Administration » getting many audit log entries from crond » 2019-01-31 17:28:54

Hmm, that does seem abberant now you present both of them.  It seems to be the 32-bit kernel is spitting out these extra log lines:

an 31 01:25:01 auditor audit[401]: USER_ACCT pid=401 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:accounting grantors=pam_access,pam_unix,pam_time acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor audit[401]: CRED_ACQ pid=401 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:setcred grantors=pam_unix,pam_env acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor audit[401]: USER_START pid=401 uid=0 auid=0 ses=14 msg='op=PAM:session_open grantors=pam_loginuid,pam_limits,pam_unix acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor kernel: audit: type=1101 audit(1548919501.573:116): pid=401 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:accounting grantors=pam_access,pam_unix,pam_time acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor kernel: audit: type=1103 audit(1548919501.573:117): pid=401 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:setcred grantors=pam_unix,pam_env acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor kernel: audit: type=1105 audit(1548919501.573:119): pid=401 uid=0 auid=0 ses=14 msg='op=PAM:session_open grantors=pam_loginuid,pam_limits,pam_unix acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor audit[401]: CRED_REFR pid=401 uid=0 auid=0 ses=14 msg='op=PAM:setcred grantors=pam_unix,pam_env acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor kernel: audit: type=1110 audit(1548919501.576:120): pid=401 uid=0 auid=0 ses=14 msg='op=PAM:setcred grantors=pam_unix,pam_env acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor audit[401]: CRED_DISP pid=401 uid=0 auid=0 ses=14 msg='op=PAM:setcred grantors=pam_unix,pam_env acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor kernel: audit: type=1104 audit(1548919501.580:121): pid=401 uid=0 auid=0 ses=14 msg='op=PAM:setcred grantors=pam_unix,pam_env acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor audit[401]: USER_END pid=401 uid=0 auid=0 ses=14 msg='op=PAM:session_close grantors=pam_loginuid,pam_limits,pam_unix acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'
Jan 31 01:25:01 auditor kernel: audit: type=1106 audit(1548919501.583:122): pid=401 uid=0 auid=0 ses=14 msg='op=PAM:session_close grantors=pam_loginuid,pam_limits,pam_unix acct="root" exe="/usr/bin/crond" hostname=? addr=? terminal=cron res=success'

And for reference, these lines (taken from the 32-bit listings) are effectively common:

Jan 31 01:25:01 auditor crond[401]: pam_unix(crond:session): session opened for user root by (uid=0)
Jan 31 01:25:01 auditor kernel: audit: type=1006 audit(1548919501.573:118): pid=401 uid=0 old-auid=4294967295 auid=0 tty=(none) old-ses=4294967295 ses=14 res=1
Jan 31 01:25:01 auditor CROND[402]: (root) CMD (touch /root/cron.touch)
Jan 31 01:25:01 auditor CROND[401]: pam_unix(crond:session): session closed for user root

I'll just note that now it's not trying to log in as you, although it does seem to be spawning a separate (root) session to do its work in still.   This appears to be a change in your configuration, but I think we can cope with that.

#20 Re: System Administration » getting many audit log entries from crond » 2019-01-31 05:54:38

Yes, those log lines all look like PAM lines to me; the root level crond is logging on as you to run your command.  if you run it locally, you don't have to go through any of those hoops.

I've no idea why this didn't used to end up in your logs; maybe the format of the messaging has changed, and your rules need updating or adding to.  But it seems to me the kind of stuff that absolutely should get logged so that if you do care about it, you can find it out.

#21 Re: Installation » Advice for partitioning a new 1 TB hard disk » 2019-01-25 17:07:27

Regarding swap, as a linux only user, I've never found a reason to create a partition and only ever use a swap *file* if doing a big compile.  I didn't realise you needed swap to hibernate to disk (although did idly wonder where it was trying to dump ram to in that case), so maybe I'll try enabling my swap file before hibernating too.  As a file, you can make it any size you deem sensible, and change it afterwards if it turns out to be too big or too small.  128MB of RAM used to be plenty to bring my old Pentium 3 running arch up from cold, and I doubt that's changed, so you don't need extra ram to boot normally.

#22 Re: System Administration » [SOLVED] After last update VLC won't start anymore: skins2 interface.. » 2019-01-25 16:57:35

On the other hand, if this was stopping vlc from loading and now works, maybe why symbols stop it loading are rather irrelevant.

Hmm, running readelf -a on my unstripped copy of libQt5Core.so gives me:

Displaying notes found in: .note.ABI-tag
  Owner                 Data size	Description
  GNU                  0x00000010	NT_GNU_ABI_TAG (ABI version tag)
    OS: Linux, ABI: 4.11.0

This seems unusual.  Not all of my libs even have a .note.ABI-tag section, but all of those that do decode to OS: Linux, ABI: 3.2.0.  Was there an ABI break in 4.11 that got reverted in newer kernels or something?

#23 Re: Installation » Advice for partitioning a new 1 TB hard disk » 2019-01-25 00:57:28

anselm wrote:

Is that the right link?  I don't see any recommendations for partition sizes there myself.  Presumably you're following an install guide (such as that one) for your other questions though.

#24 Re: System Administration » [SOLVED] After last update VLC won't start anymore: skins2 interface.. » 2019-01-24 23:32:35

I've no idea off the top of my experience why stripping symbols should make a library load; they're supposed to be orthagonal to any kind of binary load.  The superuser link seems to suggest microsoft have problems with it when loading it under WSL, but frankly their problems with loading librararies are entirely distinct from my interests.

#25 Re: Pacman / Pacman Upgrades » [SOLVED] pacman update broke postfix » 2019-01-24 23:27:09

What is courier?  It's not a package or within a package name according to pacman, and as itself I only recognise it as a (microsoft?) typeface personally.

I'll just note the OP had an error complaining about libmysqlclient.so.18 - perhaps this is part of courier, I have no way of knowing.

Board footer

Powered by FluxBB