Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
ToxicFrog
Apr 26, 2008


If it's ext2-compatible there's the ext2 IFS.

If not, there's ext2fsd, which claims limited ext4 support.

You could also install linux in a VM, mount the ext4 filesystem(s) inside that, and mount them in windows as networked drives.

Adbot
ADBOT LOVES YOU

TheGopher
Sep 7, 2009

Sizzlechest posted:

A google search tells me the best way to do it? Because if that's what you just did to come up with that recommendation, please stop posting.

No, but if you said "what's the best way" instead of saying, "I'm reading you can use $method_a or $method_b, but I don't know if those will work. Anybody know the best way?"

The second way shows you did at least a ten second google search. The way you asked the question is just begging for hand-holding. You couldn't have spent more than 20 seconds writing that post, so you're going to tell people to gently caress off when they don't spend more than 20 seconds replying to yours?

Shazzner
Feb 9, 2004

HAPPY GAMES ONLY

Who cares about hand-holding. I will hold anyones hand; for any reason. I want to touch you :h:

pram
Jun 10, 2001
Not Linux but what do people think of Solaris 11 so far? Seems a lot less masochistic than 10 was. I really like the image packaging system.

spiritual bypass
Feb 19, 2008

Grimey Drawer
Please elaborate, I'm interested.

FISHMANPET
Mar 3, 2007

Sweet 'N Sour
Can't
Melt
Steel Beams
Don't worry, there's still plenty of masochism.

I'm wondering how long before Oracle just kicks Gnome to the curb entirely and makes it server only. The only benefit Solaris has on the desktop, the ZFS based time machine, is completely out weighed by how little software there is for the platform, comparably.

pram
Jun 10, 2001

rt4 posted:

Please elaborate, I'm interested.

On what, IPS? Its essentially just a native yum/apt clone. Its pretty nice compared to Solaris package management before though, which was basically nothing at all.

Makes me wish IBM would do something similar with installp on AIX. As it stands you just use RPM because its so completely useless. Why are UNIX systems so universally awful?

Sizzlechest
May 7, 2007

TheGopher posted:

No, but if you said "what's the best way" instead of saying, "I'm reading you can use $method_a or $method_b, but I don't know if those will work. Anybody know the best way?"

I didn't realize it was my responsibility to word requests in specific ways to ensure you don't jump to conclusions that make you look like an rear end in a top hat. I apologize for not explaining each and every step that I did prior to asking for help. I now understand how that's totally your business, that complete disclosure is necessary, and how it affects you greatly in your life, or lack thereof.

JHVH-1
Jun 28, 2002

FISHMANPET posted:

Don't worry, there's still plenty of masochism.

I'm wondering how long before Oracle just kicks Gnome to the curb entirely and makes it server only. The only benefit Solaris has on the desktop, the ZFS based time machine, is completely out weighed by how little software there is for the platform, comparably.

What about things like scientific applications? Do people still use Solaris for that? It used to be the norm in the late 90s but I noticed my college switched to windows machines when they built a new lab.

FISHMANPET
Mar 3, 2007

Sweet 'N Sour
Can't
Melt
Steel Beams

JHVH-1 posted:

What about things like scientific applications? Do people still use Solaris for that? It used to be the norm in the late 90s but I noticed my college switched to windows machines when they built a new lab.

Well, Mathworks never made a version of Matlab for x86 Solaris, and they recently discontinued support for SPARC Solaris, if that tells you anything.

Harokey
Jun 12, 2003

Memory is RAM! Oh dear!

JHVH-1 posted:

What about things like scientific applications? Do people still use Solaris for that? It used to be the norm in the late 90s but I noticed my college switched to windows machines when they built a new lab.

I can't speak for everything, but I personally migrated a few systems a couple years ago from Solaris to Linux because the software vendor stopped supporting solaris on their newest software releases.

FISHMANPET
Mar 3, 2007

Sweet 'N Sour
Can't
Melt
Steel Beams

FISHMANPET posted:

Well, Mathworks never made a version of Matlab for x86 Solaris, and they recently discontinued support for SPARC Solaris, if that tells you anything.

And I'm going to double post instead of quote, so nobody misses this:

Solaris 10 Desktop is awful. Truly awful. I can say this having used it for quite a while, as my school supports it in the labs. For Solaris 10, Sun said "Hey, lets fork Gnome!" So they did that. And then never changed anything ever again. Through all the Solaris 10 updates, Gnome stayed exactly the same. They never improved it, they never added anything from upstream, they just left it as a lovely XP clone called JDS. Coupled with the fact that it's a pain in the rear end to develop and test on Solaris (since it's a non-free Unix) most of your standard Desktop stuff isn't going to work (VLC, Rhythmbox come to mind). And since it's bastardized Gnome that's never been updated, your stuck with horribly out of date GTK libraries, so compiling any modern GTK application is a nightmare.

Apparently the SPARC architecture is awesome, which was frequently bandied about as a reason we supported Solaris, but it was only awesome on the high end. There's no reason to fill a room with lovely SPARC workstations when you can spend half as much for twice as much power with Linux, and spend the leftovers on a beefy Sparc server to do your actual calculations. On top of that, Sun hasn't made a SPARC workstation for years, so you can't run this decades long pedigree of Sun software.

And on top of that, Solaris 10 always ran like poo poo on SPARC. When I described it to our Solaris admin, I told him that using a SPARC workstation was like using a Windows install with no graphics driver and everything done on the CPU. Maybe before Solaris 10 it was useable, but once 10 came around the Sun Desktop team (if it existed) decided to spend about 20 minutes writing some code, and spent the next 6 years with their fingers in their ears screaming "LALALALALALALA"

OpenSolaris made some progress, as they'd fork Gnome for every release, and then add their changes, so porting was a lot easier. There are a lot of Solaris zealots who have to run Solaris on their personal machines, but I've no idea why. There are plenty more who use Macs and SSH to their Solaris machines, which makes way more sense.

We'll see what Oracle does with Solaris 11 Desktop, but I'm hoping they get rid of it. They've taken the first step, and released Solaris 11 Express with a text only installer (Open Solaris never had a text only installer, you installed from within a LiveCD Session).

The tl;dr version I'm mad about being forced to use Solaris Desktop for years because it's such an awful experience, and I can't for the life of me fathom any use of Solaris on the desktop that isn't made easier by a Linux/Mac/Windows workstation connecting to a Solaris server.

And as a disclaimer, I run Solaris 11 Express on my home server and I absolutely love it. I love Solaris on the server, and I'm pretty mad that Oracle was a dick about OpenSolaris.

covener
Jan 10, 2004

You know, for kids!

Pram posted:

Makes me wish IBM would do something similar with installp on AIX. As it stands you just use RPM because its so completely useless. Why are UNIX systems so universally awful?

I always assumed with actual education that 'suma' on AIX was useful, but the online writeups are not so easy to grok from a casual administrators perspective.

pram
Jun 10, 2001

covener posted:

I always assumed with actual education that 'suma' on AIX was useful, but the online writeups are not so easy to grok from a casual administrators perspective.
suma isn't close to being as useful as ips or yum, I don't even think you can roll your own repository. Gonna be honest I've never looked into it though since package management on AIX is such a clusterfuck and the extent of my experience with suma is pressing update butan in smit.

hackedaccount
Sep 28, 2009

JHVH-1 posted:

What about things like scientific applications? Do people still use Solaris for that? It used to be the norm in the late 90s but I noticed my college switched to windows machines when they built a new lab.

Back in the early 80's when Sun first started they were the poo poo because they were (relatively) cheap for their functionality. Once they developed SPARC it was very fast, but expensive compared to x86 stuff. For a good decade or so if you needed pure horsepower, RISC (SPARC, PA-RISC, POWER, MIPS) was the only way to go - x86 just wasn't fast enough. Then, x86 speeds increased dramatically and of course were a lot cheaper than RISC stuff - it was the end of the line for RISC on the desktop. Also, instead of a company writing a version of their software for AIX, one for Solaris, one for HP/UX, and one for IRIX, they could just write a Windows version and a Linux version and be done with it.

Solaris has always been a pretty crappy desktop OS, but people had to put up with it in order to get access to the blazing fast processor underneath.

Vulture Culture
Jul 14, 2003

I was never enjoying it. I only eat it for the nutrients.

Accipiter posted:

An ext4 filesystem should be backwards-compatible with ext3 and 2, so a reader like ext2read should be able to handle it.
Yeah, this is almost completely wrong.

An ext4 filesystem that's been upgraded from ext3 does not use extents, and will be mostly-readable with an ext2/ext3 utility. An ext4 filesystem that's been created from scratch with extents uses a completely different underlying format from ext2/ext3.

Accipiter posted:

...but you'd have known that if you could have bothered to do a simple Google search for ext4 windows.
You could stand to Google "ext4 ext3 backward compatibility" yourself.

JHVH-1
Jun 28, 2002

hackedaccount posted:

Back in the early 80's when Sun first started they were the poo poo because they were (relatively) cheap for their functionality. Once they developed SPARC it was very fast, but expensive compared to x86 stuff. For a good decade or so if you needed pure horsepower, RISC (SPARC, PA-RISC, POWER, MIPS) was the only way to go - x86 just wasn't fast enough. Then, x86 speeds increased dramatically and of course were a lot cheaper than RISC stuff - it was the end of the line for RISC on the desktop. Also, instead of a company writing a version of their software for AIX, one for Solaris, one for HP/UX, and one for IRIX, they could just write a Windows version and a Linux version and be done with it.

Solaris has always been a pretty crappy desktop OS, but people had to put up with it in order to get access to the blazing fast processor underneath.

I remember using those machines when I first started in college and they were sweet to use. I was just getting into Linux as my desktop so I would go into the engineering lab and use CDE/Netscape to check mail etc.
Later on they put Gnome on there as an option but because they were using AFS for certain software or something it was a mess and half the things didn't work. That was around the time they started getting windows machines too. Dual core or dual cpu processors were expensive or pretty rare back then so they were pretty impressive at first. A lot of people hadn't even heard of Linux when I was using it as my desktop, but after people like IBM, HP got into it and companies doing Linux software could become public on the stock market it started to take off. Its crazy now my telephone runs linux and is faster than the first machine I ran it on ( a Cyrix 6x86 overclocked to a whopping 166Mhz).

Accipiter
Jan 24, 2004

SINATRA.

Sizzlechest posted:

I apologize for not explaining each and every step that I did prior to asking for help.

...except that SHSC rules tell you to do EXACTLY THAT. Imagine!

Misogynist posted:

Yeah, this is almost completely wrong.

An ext4 filesystem that's been upgraded from ext3 does not use extents, and will be mostly-readable with an ext2/ext3 utility. An ext4 filesystem that's been created from scratch with extents uses a completely different underlying format from ext2/ext3.

While I will admit not being too terribly familiar with ext4's backward compatibility (and that is why I said "should be backwards compatible" and not "is backwards compatible", Ext2Read supports ext4's extents and I THINK Linux Reader does as well.

dont skimp on the shrimp
Apr 23, 2008

:coffee:

Accipiter posted:

While I will admit not being too terribly familiar with ext4's backward compatibility (and that is why I said "should be backwards compatible" and not "is backwards compatible", Ext2Read supports ext4's extents and I THINK Linux Reader does as well.
That's not what you said though.

Thing is, ext4 isn't usually mountable as ext3/2. You need ext4 support to mount it at all. You can add some flags to mkfs to make it fully compatible with ext3/2, but you lose most of what makes it a good fs.

If ext2read has support for ext4, then yeah, that's what you should've said from the start. What you did say was just plain wrong though.

Accipiter
Jan 24, 2004

SINATRA.

Zom Aur posted:

That's not what you said though.

Um... Really?

Accipiter posted:

An ext4 filesystem should be backwards-compatible with ext3 and 2, so a reader like ext2read should be able to handle it.
Incidentally that happens to be exactly what I said.

Zom Aur posted:

If ext2read has support for ext4, then yeah, that's what you should've said from the start.

Yeah, that I'll agree with. But that IS why I suggested it.

mike12345
Jul 14, 2008

"Whether the Earth was created in 7 days, or 7 actual eras, I'm not sure we'll ever be able to answer that. It's one of the great mysteries."





I'm looking for someone to write a script for my website, more info @ http://forums.somethingawful.com/showthread.php?threadid=3398004 (SA Mart).

Basically the scripts needs to crawl folders of static html and randomize links and images. I pay 20 bucks.

Bob Morales
Aug 18, 2006


Just wear the fucking mask, Bob

I don't care how many people I probably infected with COVID-19 while refusing to wear a mask, my comfort is far more important than the health and safety of everyone around me!

What could make this server think it has 5 CPU's? It should have 8 (2 x quad-core)

code:
[deploy@275368-app1 current]$ cat /proc/cpuinfo
processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4521.99
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]

processor	: 1
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4521.85
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]

processor	: 2
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4522.20
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]

processor	: 3
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4521.71
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]

processor	: 4
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4521.83
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]
??

edit:

$ uname -a
Linux 2.6.18-194.17.1.el5 #1 SMP Mon Sep 20 07:12:06

(CentOS)

Bob Morales fucked around with this message at 21:33 on Mar 18, 2011

dont skimp on the shrimp
Apr 23, 2008

:coffee:

Accipiter posted:

Um... Really?

Incidentally that happens to be exactly what I said.
You said any reader should work if it has support for ext2/3, you didn't specify that those two readers should work because they specifically has support for the newer features of ext4.

This is what I meant:

Accipiter posted:

Ext2Read supports ext4's extents and I THINK Linux Reader does as well.



Anyway, to not be completely off topic: I just gave systemd a go on my netbook. It's incredably fast, though you might have to create some scripts to launch certain daemons (pretty simple formatting though).

Anyone else using this?

Morax
Feb 26, 2011
I have a problem with ubuntu. If anyone can help. I have 10.10 on a disc and installed it (dual boot). I've asked on ubuntu forums, an ubuntu mirc channel, and yahoo answers. So far, no one knows what to do. My mouse pointer only clicks on a link, internet or otherwise, when you point it at the high end of the link. I'm going to try and reinstall, but has anyone had this problem before? Any type of mouse settings that might fix this?

My system is a hp d325 amd xp 2.0 ghz, with xp on the other boot. I am thinking it might be my outdated computer, but maybe not.

dont skimp on the shrimp
Apr 23, 2008

:coffee:

Morax posted:

I have a problem with ubuntu. If anyone can help. I have 10.10 on a disc and installed it (dual boot). I've asked on ubuntu forums, an ubuntu mirc channel, and yahoo answers. So far, no one knows what to do. My mouse pointer only clicks on a link, internet or otherwise, when you point it at the high end of the link. I'm going to try and reinstall, but has anyone had this problem before? Any type of mouse settings that might fix this?

My system is a hp d325 amd xp 2.0 ghz, with xp on the other boot. I am thinking it might be my outdated computer, but maybe not.
Does the problem persist even in a different browser?

Morax
Feb 26, 2011

Zom Aur posted:

Does the problem persist even in a different browser?



It's all around the operating system, everything.

darkhand
Jan 18, 2010

This beard just won't do!

Bob Morales posted:

What could make this server think it has 5 CPU's? It should have 8 (2 x quad-core)

[code][deploy@275368-app1 current]$ cat /proc/cpuinfo
<snip>
$ uname -a
Linux 2.6.18-194.17.1.el5 #1 SMP Mon Sep 20 07:12:06

(CentOS)

Bad bios-flash? I don't think linux-acpi can do low-level enough stuff like turning off the cores, but I could be wrong. Even if it did, it seems like /proc/cpuinfo would still pick it up

darkhand fucked around with this message at 18:32 on Mar 19, 2011

JHVH-1
Jun 28, 2002

Bob Morales posted:

What could make this server think it has 5 CPU's? It should have 8 (2 x quad-core)

code:
[deploy@275368-app1 current]$ cat /proc/cpuinfo
processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4521.99
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]

processor	: 1
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4521.85
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]

processor	: 2
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4522.20
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]

processor	: 3
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4521.71
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]

processor	: 4
vendor_id	: GenuineIntel
cpu family	: 6
model		: 26
model name	: Intel(R) Xeon(R) CPU           L5520  @ 2.27GHz
stepping	: 5
cpu MHz		: 2260.999
cache size	: 8192 KB
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi
 mmx fxsr sse sse2 ss syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm
bogomips	: 4521.83
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: [8]
??

edit:

$ uname -a
Linux 2.6.18-194.17.1.el5 #1 SMP Mon Sep 20 07:12:06

(CentOS)

Never seen this happen before. I would toss a linux live cd like riplinux in and see what it reports.

TheGopher
Sep 7, 2009
I don't know if anybody here has any experience with Cassandra, but I'm getting pretty frustrated. I've been messing around with some stuff and decided to set up a Reddit clone and hopefully learn something in the process.

On my Debian 6 server, I can get the Cassandra daemon to run just fine. I can connect to the instance with casssandra-cli, but for some reason it isn't loading the keyspace defined in /etc/cassandra/cassandra.yaml. I'm really not sure what I'm supposed to do next, because from my understanding that's all I need to do. I've been searching around, and I saw a reference to LiveSchemaUpdates, but I can't find anything about why the keyspace isn't important from the YAML file.

My Rhythmic Crotch
Jan 13, 2011

Ugh, I'm having some really lovely video issues. I have openSuse 11.4 installed, and I'm using a Quadro NVS 285 GPU. Everything was fine for an hour or so, then of nowhere I started getting snow, stars, black bars, lock ups, etc. KDE becomes completely loving unresponsive during these episodes. I can still SSH in and reboot the machine, but the issues start again after rebooting.

I have a spare machine at work that has the same version of openSuse, same GPU, same driver setup, and it runs like a champ.

I guess I just got a bunk video card, right? :(

(edit for clarity: the NVS 285 is the new addition to the mix here)

My Rhythmic Crotch fucked around with this message at 23:52 on Mar 19, 2011

covener
Jan 10, 2004

You know, for kids!

hootimus posted:

Ugh, I'm having some really lovely video issues.

I vaguely recall battling a similar issue forever ago by explicitly telling the X server how much video memory there was, but now that there's usually not any/much X server config this is probably a longshot.

My Rhythmic Crotch
Jan 13, 2011

The strange thing is that the GUI seems to run fine in an NX session even when the local display is hosed. If NX utilizes the GPU (I think it does, right?) that could mean ... well I don't know wtf that could mean. Why would it work for NX but not for the local display?

I guess I could put the GPU in my windows machine and see if it's stable there. Ugh.

smug forum asshole
Jan 15, 2005
On the off chance that any of you are command-line geeks who don't already use GNU Screen: I wrote a little primer on screen yesterday and I'd like to get it out there. I hope it's helpful to someone :hehe:

Bob Morales
Aug 18, 2006


Just wear the fucking mask, Bob

I don't care how many people I probably infected with COVID-19 while refusing to wear a mask, my comfort is far more important than the health and safety of everyone around me!

JHVH-1 posted:

Never seen this happen before. I would toss a linux live cd like riplinux in and see what it reports.

It's co-located, and it's in production so I can't do that. I will probably check with Rackspace and see if they can check the BIOS during the next maintenance time.

I didn't think a BIOS would let you disable cores on a per-CPU basis but who knows.

Vulture Culture
Jul 14, 2003

I was never enjoying it. I only eat it for the nutrients.

Bob Morales posted:

It's co-located, and it's in production so I can't do that. I will probably check with Rackspace and see if they can check the BIOS during the next maintenance time.

I didn't think a BIOS would let you disable cores on a per-CPU basis but who knows.
How many CPUs do your other utilities like top and sar show?

(There are several kernel threads that are instantiated once per CPU. That's what you're counting in top.)

Ninja Rope
Oct 22, 2005

Wee.
What does dmesg from boot say? Anything interesting about offlining CPUs?

RHEL5 dmesg posted:

Total of 2 processors activated (8513.27 BogoMIPS).
[...]
Brought up 2 CPUs

TheGopher
Sep 7, 2009

TheGopher posted:

I don't know if anybody here has any experience with Cassandra, but I'm getting pretty frustrated. I've been messing around with some stuff and decided to set up a Reddit clone and hopefully learn something in the process.

On my Debian 6 server, I can get the Cassandra daemon to run just fine. I can connect to the instance with casssandra-cli, but for some reason it isn't loading the keyspace defined in /etc/cassandra/cassandra.yaml. I'm really not sure what I'm supposed to do next, because from my understanding that's all I need to do. I've been searching around, and I saw a reference to LiveSchemaUpdates, but I can't find anything about why the keyspace isn't important from the YAML file.

I figured this out. Issue was I thought schematool HOST PORT import uses JMX, not RMI, and as such needs to connect on port 8080, not 9160...

Yeah I don't really know what the gently caress I'm doing. At least I'm learning something about Cassandra...

ToxicFrog
Apr 26, 2008


hootimus posted:

The strange thing is that the GUI seems to run fine in an NX session even when the local display is hosed. If NX utilizes the GPU (I think it does, right?) that could mean ... well I don't know wtf that could mean. Why would it work for NX but not for the local display?

By this I assume you mean if you log into the system having problems using NX, everything looks fine? This is expected. The server's GPU is not at all involved in the process; it just sends drawing instructions to the client.

What this sounds like is the GPU overheating. The new KDE drives the video card very hard; I'm guessing that the GPU in the other machine can handle it, but this one is below spec and starts to have issues much earlier than it should.

dolicf
Sep 12, 2010

Bob Morales posted:

Oh no, 5 cores!

I'm guessing this is a virtual machine and therein lies the root of the issue. Have Rackspace check the hypervisor and make sure that the the host is allocated the appropriate number of vCPUs. I'd put money on this if I were a betting man.

E: But seriously this is the case. Trust me on this one.

dolicf fucked around with this message at 11:35 on Mar 20, 2011

Adbot
ADBOT LOVES YOU

My Rhythmic Crotch
Jan 13, 2011

ToxicFrog posted:

By this I assume you mean if you log into the system having problems using NX, everything looks fine? This is expected. The server's GPU is not at all involved in the process; it just sends drawing instructions to the client.

What this sounds like is the GPU overheating. The new KDE drives the video card very hard; I'm guessing that the GPU in the other machine can handle it, but this one is below spec and starts to have issues much earlier than it should.
I think you're right. The GPU in the stable machine is fan cooled but the one in the machine with issues is passively cooled... that must be the problem.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply