Fedora :: What Is The Timing On Updated RPMs For New BOINC Versions
Jun 4, 2010
I've just recently started using Fedora again and have a question about BOINC.I see that Fedora 13 is using 6.10.45, but the latest recommended release is 6.10.56.What is the timing on updated RPMs for new BOINC versions? Or is there a guide to updating manually, which I've been unable to locate?
Trying to get the thunderbird localization working I briefly checked thunderbird 3, to see if it had the same issue. Now, the problem is, I accidentally installed new versions of Firefox 3.5 and XULrunner. If I remove them, then many other packets get removes due to dependencies. Same if I try to force version. If I simply remove the repos, then synaptic reports them as broken. Is there anything I can do to solve this? How stable are the updated versions? How often do they get updated?
I have Fedora 15 AMD64 installed on a AMD Athlon II X2 250 3.0Ghz, 8GB DDR3 1333Mhz, with a ATI Radeon HD5450 Graphics card recently installed. Problem is that BOINC won't recognize the GPU. I know the GPU is working because I've downloaded some MKV files from VODO, and I can play them while running BOINC at a fairly high level and they play fine (using Videolan as the player). Before I installed the GPU playback on the same files was impossible to watch, it was extremely jerky.
So now I need to get BOINC to use the AMD GPU, so I can upgrade my computation speed. And I need to get it set so that if I decide to do something that needs the GPU, that it releases it. I'm considering installing Neverwinter Nights on this computer. I never did finish the game before I dumped Windows
I've installed the Boinc 6.4.5 x86_64 using F10 update package from Fedora repo. The client runs fine using the CPUs, but doesn't fing the nVidia display processors. I've installed nVidia drivers using Fusion repo packages for F10 ( currently 180.25 ) nVidia display settings seems happy and correctly reports driver 180.25 in place as well as 9600 GSO graphics card and LCD monitor.
My objective is to get Boinc using the CUDA capabilities of my GeForce 9600 GSO graphics card and see some of the performance benefits advertised. Although Boinc client is running it will only use the CPU to do work. I have not been able to discover any flags or configuration settings to help Boinc find CUDA library files. I'm stuck at the limits of my knowledge. The following are the startup messages from Boinc. I'm assuming the fact that it can't find the library causes the inability to see the GPU coprocessor.
I am very new to linux and I am running into an issue..I've gotten soem new HP blade servers and they cam installed with RHEL 5.3 I need to update them to RHEL 5.4 I register them with my rhn subscription, then once I see my systems online, i run yum update from the command line.
It seems to find a ton of updates and takes a long while to DL everything.Do I have to tell it to install the downloads? I don't think I do...I get some errors about timing out for the last few updates.theni reboot, but when i check redhat-release it still says 5.3
A few weeks ago I upgraded one of my laptop from Fedora 13 to Fedora 14 (using pre-upgrade). The upgrade went smoothly and no errors appeared. However ever since the upgrade ssh is not working anymore. I've tried various servers inside and outside the local network. I can't contact any of them. In all cases I get:
Code: ssh: connect to host aaa.bbb.ccc port 22: Connection timed out I've tried to get more info using the -vvv option form ssh, but it doesn't mean too much to me:
Code: OpenSSH_5.5p1, OpenSSL 1.0.0d-fips 8 Feb 2011 debug1: Reading configuration data /etc/ssh/ssh_config debug1: Applying options for * debug2: ssh_connect: needpriv 0 debug1: Connecting to aaa.bbb.ccc [xxx.xxx.xxx.xxx] port 22. debug1: connect to address xxx.xxx.xxx.xxx port 22: Connection timed out ssh: connect to host aaa.bbb.ccc port 22: Connection timed out
Actions so far: - checked internet connection. No problem there. Another laptop is working fine using the same connection and the same ethernet cable. That laptop is still using Fedora 13, since I didn't want to get stuck without ssh completely. - checked "messages" log. No messages at the time of ssh connect attempt. - checked "secure" log. No messages at the time of ssh connect attempt. - checked the firewall settings. The settings are exactly the same as before the upgrade (when ssh was still working). Moreover, the settings are the same as the other laptop that is working. - temporarily switched off the firewall. No difference. - temporarily switched off selinux. No difference.
So it doesn't seem to be a firewall or selinux problem. And I know the connection is working, so it does not seem to be a routing problem either. What am I missing here?
System specs: - Asus EEE PC (1 GB RAM, intel atom processor); - Fully up-to-date Fedora 14 (kernel 2.6.35.13-91.fc14.i686)
Running a LAMP server, CentOS as the OS.The sites always been slow, but now that ive optimized it with mysql cache, gzip compression and some other things, its really fast.Except when pages loading seem to randomly 'time out'. The browser sits on 'waiting for x.com'. Closing the browser and/or the tab and opening a new one fixes it, but then it'll happen again eventually. Clicking further links while its 'waiting for x.com' does nothing, basically the site becomes unusable until you close the tab and reopen it.
This happens on all 3 virtual servers we're running within apache. Mainly noticable on the PHPbb forums, probably because they are visited the most.It's not a slow mysql query, i turned on slow query logging over 2 seconds, and the only two hits i got on that i know are unrelated.Ive turned off some optimizations thinking they might be it, but no dice.
On formula1.com live timing which is java applet starts but does not refresh. Time of training is not counting and there are no new sector times and overall times for drivers laps. Firefox is 3.6.3 and jre is jre-1.6.0_20-fcs.i586.
Anybody know of any Firefox 3.5 final RPMS for Fedora 10? I looked a rebuilding the Fedora 11 RPMS for Fedora 10, but the other packages requiring an upgrade as well is getting out of control.
I'd like to confirm if the latest updates of a couple of rpms is causing me a problem. I'd like to downgrade these rpms to an intermediate version (greater than the original F11 versions, but less than the latest version). I've looked on a few mirrors, and all I can find is the latest update. Are intermediate versions of an rpm discarded when they're replaced with new versions? Are these intermediate versions available some place?
The two rpms I'm interested in are: gstreamer-plugins-good-0.10.16-1.fc11.i586 gstreamer-plugins-base-0.10.25-1.fc11.i586
I am trying to create a local repository of installed RPMs on my workstation. Using the 'rpm -qa' I can gather the installed RPMs on my workstation. Is there a way for me download all these RPMs using the output of 'rpm -qa'? 'yum --downloadonly' will allow me the option of downloading without installing. Since the installed RPMs are not cached in my workstation I have to download these RPMs again. If I have a local repo then it will be easier for me restore my installation without having to rediscover the packages needed. I am looking for how I can download multiple RPMs.
I have noticed that the updates to my f12 64bit systems are running longer due to the fact that fewer fedora updates are delta rpms. Why is this. I know that rpmfusion updates and other third party updates are not covered.
I know there is a command that will show what rpms are installed on a machine that are no longer available in a yum repository, but I can't remember what it is! It is useful to identify packages left over after an upgrade that are no longer needed, or to identify third party software.
I have been running f13 ever since it was released. Evidently the kmod-rt2870 rpms are never deleted when older kernels are removed. When I tried to do today's update the Package Manager and yum complained that some of the older kmod-rt2870 packages left over from fedora 12 were missing kernel dependencies. The kernels have been missing for a long time. I don't know why the package manager started complaining today.
I tried 1) yum clean all 2 rpm --rebuilddb 3) removed the /lib/modules directory corresponding to the old kernels and the problem persisted.
I finally removed all of the kmod-rt2870 rpms what where originally installed in fedora 12 and I was finally able to complete the update. Why did the package manager suddenly start complaining about the missing kernels?
I just installed F14 (so far I like it ). I had F13 but decided to wipe the old OS and install F14 fresh. In the course of replacing all of my old programs, I had to download several .rpms like the .rpms for the RPMFusion repos and the .rpm for Adobe Reader. After downloading the .rpms to a download directory I ran yum localinstall to install them. The installations went well and I didn't have any problems.
However, now the packages are showing up as updates which need to be installed in the software updates window. I also ran yum check-update and they are showing up as updates which need to be installed. I tried deleting the .rpms and also running yum clean all but they are still showing up as updates which need to be installed.
Using yum installs the binary version of the software. Is it possible to use yum to download the src rpms , because i have seen one repo called Fedora src. I am using Fedora 11.
I am trying to install f10 on an old dell dimension 2400 with an old video chipset. Everything was working pretty smooth until I tried installing an .rpm file from a flashdrive. At first, it looked like it was installing the rhpxl file but gave a warning message that said
warning:rpmts_HdrFromFdno:Header V3 DSA signature: NOKEY, Key ID 4ebfc273
then says the public key will not be installed. Is there a file/command that I can download and run from a flash because I have no capable internet connection because for some reason linux doesn't see it in the install.
Okay, according to CUPS, I've installed this printer, but whenever I try to print, it tells me it's completed, but the printer does nothing. So I search a little and found some Canon Linux drivers from Canon Europe. I downloaded the RPM version, which gave me two files: cnijfilter-common-3.00-1.i386.rpm and cnijfilter-ip3600series-3.00-1.i386.rpm. So I installed the common file, followed by the ip3600 file, using rpm -ivh [file name] at command line. The RPM install went fine. However, what do I do now? I tried searching the CUPS list of printers, and nothing new is listed (oddly, CUPS seemed to have drivers for virtually every iPxxxx printer except the 3600 when I first installed). I tried restarting cups, no difference. I even tried to print (suspecting nothing had changed, but hoping) and it still thinks it's printing, but nothing comes out. What did installing those two RPMs do? Did it put a file somewhere that I need to reference in the cups admin page?
Have project to move to RHEL5 with Oracle11g. During process, went to Red Hat Network and they were not currently available. Is there another source of RPM's for installation on RHEL5?
I am trying to set up my cpu freq. scaling to ignore BOINC. From what google has shown me this is done by setting a value of 1 for ignore_nice_load. However the location of said value does not seem to be the same in 10.04 as in the results from google. How do i set this or is there a better way to keep idle processes like BOINC from increasing the my cpu frequency?
I am running Testing with the liquorix kernel. Right now I have Boinc 6.10.52 on here. I have tried 6.10.58, 6.12.8 and 6.12.18 (all for 64bit linux). I have, for months, been running the 6.12.8 version. Boinc is installed in my home folder as it has been for nearly 2 years.
Rebooted after trying to test an Ubuntu LiveCD for the 11.04B1 (Failed big time). I turned off this drive in bios for that test. This has been done many times over the years of testing as I use a different drive for that and want no contamination of my main drive. I did have some problems booting that were corrected by reinstalling my fglrx-driver package.
The only problem is that Boinc will not run at all. None of those versions. They all behave the same way. Open boincmngr from menu and it comes up blank and "disconnected".Go to Advanced/Select computer. Has the option for "Debian". This is correct. Select that and enter the password for Boinc or the password for Debian and it tries to connect but does not.
I have a home network with 6+ x86_64 machines, all with similar setups. In the past (FC10 and before), I've had common package repositories (e.g /var/cache/yum/fedora/packages) shared via NFS with all the machines (and with keepcache=1 in /etc/yum.conf). That way, a given RPM only got downloaded once; the other machines would then pull it from my local package repository. And I don't mind the disk usage of keeping one copy of all my old RPMS around.
It seems that while DRPMS is great for a single machine, it doesn't make sense in my case. If I have to download the DRPM 6 times (and take the time/CPU hit to recreate the RPM 6 times), I might as well have downloaded the RPM once and been done with it. Is there a not-too-convoluted method to keep a common package repository across multiple machines even with DRPMS? Or, better, to have that first downloading machine pull a DRPM, generate the RPM, then save the RPM in the local shared repository?
It is not difficult to set up a regular local repository (rpms) and manage a fedora system with it using yum. I wonder if the process is same for deltaRPMS. Has anyone successfully set up a local deltaRPM repo or is this something reserved for online repos? How does one achieve this?
i have a question about rpm.pbone.net, www.rpmseek.com and these kind of pages: Can I trust these sites and is it secure to install rpms from these sites on enterprise workstations and servers?
I've been using BOINC Manager to run Einstein@Home for about a week now on Ubuntu 10.04 Before I started using it I installed the Cuda 3.0 toolkit driver downloaded off of nVidia's website. BOINC ran flawlessly until today. I don't know what changed, but now the status of my tasks is "GPU Missing, Ready to Run(1.00 CPUs + 1.00 NVIDIA GPUs)
I've installed BOINC for first time (from suse repos). I'm worried about running BOINC as root. How can this be avoided? I'd first like to exhaust all options with the official opensuse repo version of BOINC. If I am unsuccessful, then I'll try the version from Berkeley website.
I'm not to clear on the difference between LTS versions and other versions, but think I may want to go with LTS. Can someone tell me if my thinking is correct given the following situation: I have some very cool, but very expensive software installed with a group license from my school, a school which I am not going to be attending for too much longer. So I want to go as long as possible without reinstalling Ubuntu, because once the product is licensed it will be licensed until I reinstall Ubuntu (or I uninstall the program). So I think this is going to require me keep the Ubuntu version I install as long as possible.
So in this case, should I go with 10.04LTS or should I just install Natty Narwhal and keep that as long as possible? It looks like 10.04LTS will be "supported" longer, but I'm not exactly clear on all that "supported" entails. Presumable it means security and software updates will be available for 10.04LTS for much longer than 11.x versions? So I'm thinking I should go with 10.04LTS
Is my thinking correct in going with 10.04LTS? Edit: It was pointed out that this would be against my contractual agreements. Which I suppose is probably true.