« first day (3653 days earlier)      last day (1316 days later) » 

1:59 PM
I got an email from a recruiter looking for a "Strong windows system admin"
I kind of want to reply with a picture of me weightlifting and say "Yo I'm strong as hell bro"
 
@jesse_b and another one showing you tweaking your X window system settings
 
hah
@StephenKitt: You like egg nog?
 
@jesse_b I don’t think I’ve ever had one!
 
@StephenKitt Is it not common in France or it just never appealed to you?
It's only available here seasonally (for the most part) it starts showing up around this time. My grocery store has had "pumpkin spice egg nog" for like a month now but they still don't have any regular stuff. Then it goes away sometime after christmas
 
@jesse_b I’ve heard of it but never run into it; in the various places I’ve lived, the “warm-you-up” winter drink is mulled wine or heated apple juice
 
2:09 PM
A quick google search tells me that in France it's typically called "hen's milk"
 
yeah I’ve only heard of that in Québec
 
I think I've had mulled wine, in Germany. It was called "gluhwein"
I probably would have to acquire the taste for it. It threw me off since I'm not used to drinking warm wine.
 
Yeah, it’s an acquired taste; most of the taste comes from the spices added to the wine.
 
On todays date in 1876 the first two way telephone call took place
 
So, systemd newer kernels have broken nice. Thoughts?
We were just bitten by this at work: I kept bugging the C++ people to compile with nice, they finally started doing so and it made no bloody difference! (Ubuntu servers, don't ask me why)
 
2:24 PM
@terdon I don’t think it’s related to systemd
 
Ah no, fair point. It's a kernel change, right?
 
@terdon yup, which has to be enabled IIRC; I’m not sure what’s enabling it now
 
Apparently, Ubuntu and Mint enable it by default and so, to my surprise, does Arch
$ cat /proc/sys/kernel/sched_autogroup_enabled
1
And so do RHEL and Manjaro according to that redit thread
 
Yup, and Fedora and Debian
 
I really think fedora would become a more popular OS if they changed the name
 
2:28 PM
But, but, why? Why would you break something as established as nice?
 
@terdon nice programs finish last
 
They're supposed to! :P
 
unintended consequence maybe?
 
Seems like a pretty major fumble to me. Certainly surprised all of us, including our sysadmin.
 
I don't know, if a kernel bug only breaks one program that probably isn't that large of a bug
 
2:34 PM
Oh it is, user-space regressions are the biggest no-no in the Linux kernel
 
yeah, someone posted a thread where Linus was going ballistic over that the other day
 
it’s how I got my first kernel patch in
 
@terdon :(
 
@terdon incidentally, tasks which are assigned a cgroup aren’t autogrouped, so systemd might help ;-)
 
Ha!
And yes, we're looking at cgroup as an alternative. I still feel it's very wrong to break nice!
 
2:49 PM
$ cat /proc/sys/kernel/sched_autogroup_enabled
0
debian ^ (mine at least)
 
Yeah, the redit thread says that Debian doesn't enable it.
But @StephenKitt said it did. Shame, Stephen, shame!
 
My main Debian 10 system has it enabled, I haven’t checked the others
 
3:10 PM
 
whaaaaa?
And it's been effectively nullifying nice since then?
 
@terdon no, it needs to be enabled in the kernel configuration (CONFIG_SCHED_AUTOGROUP), so it’s only been enabled since your distro changed that
 
3:29 PM
stupid distro
 
 
1 hour later…
4:45 PM
I installed google chrome by running dnf install on the local google chrome rpm file, and I got these errors. Does anyone know what they mean? The installation still completed despite the errors.
error: can't create transaction lock on /var/lib/rpm/.rpm.lock (Resource temporarily unavailable)
error: /tmp/google.sig.GT3eFx: key 1 import failed.
error: can't create transaction lock on /var/lib/rpm/.rpm.lock (Resource temporarily unavailable)
error: /tmp/google.sig.GT3eFx: key 2 import failed.
Redirecting to /bin/systemctl start atd.service
I'm thinking of removing the installed google chrome package because these errors might cause problems in the long run?
 
5:08 PM
What is the difference between dnf history rollback and dnf history undo?
> dnf history rollback <transaction-spec>|<package-file-spec>
Undo all transactions performed after the specified transaction. Uses the last transaction (with the highest ID) if more than one transaction for given <package-file-spec> is
found. If it is not possible to undo some transactions due to the current state of RPMDB, it will not undo any transaction.

dnf history undo <transaction-spec>|<package-file-spec>
Perform the opposite operation to all operations performed in the specified transaction. Uses the last transaction (with the highest ID) if more than one transaction for
They seem identical to me, so why are they separate commands?
 
 
2 hours later…
6:55 PM
This is probably a far out question, but can anyone think of an automated way to check for page duplicates in a PDF document?
 
 
1 hour later…
Tim
8:17 PM
Yes, split the pdf file into image per page, and use some image software to find duplicate
 
@Tim You got any eggnog in stores out there yet?
 
Tim
8:46 PM
@jesse_b church pantry doesn't have dessert
on the other hand, I have never heard of it
They once had french cookies
 
9:06 PM
@FaheemMitha how exact duplicates are we talking?
pdfseparate that comes with pdflatex(?) will blow up your pdf, then you only have to solve the page-by-page comparison. So for once I agree with what Tim said.
comparison can range from diff through simple script to heavy-duty image processing work
 

« first day (3653 days earlier)      last day (1316 days later) »