Re: [CentOS] Socket behavior change from 6.5 to 6.6
I'd like to thank everyone for their replies and advice. I'm sorry it took so long for me to respond; I took a long weekend after a long shift. Some remaining questions can be found in the final section of this posting. The summary (I hope i have all of this correct): Problem: A DOS box (client) connects to a Linux box (server) using the same local port (1025) on the client each time. The client sends data which the server reads; the server is passive and does not write any data. If the client crashes and fails to properly close the connection, under CentOS 6.5, the unclosed listener on the server receives a 0-length recv(), allowing for a clean reconnect; under 6.6, it does not, and the client unsuccessfully retries the reconnect endlessly. Diagnosis: Because the client is connecting using the same port every time, the server sees the same 5-tuple each time. At that point, the reconnection should fail until the old socket on the server is closed, and the previous behavior of receiving a 0-length recv() on the old server socket is unsupported and unreliable. Until the update to CentOS 6.6 'broke' the existing functionality, I had never looked deeply into the connection between the client and the server; it 'just worked', so I left it alone. Once it did break, I realized that because the client was connecting on the same port every time, the whole setup might have been relying on unsupported behavior. My workaround: I unfortunately had to implement an emergency workaround before receiving any replies. Fortunately, the client also sends status messages to the same computer (but a different server program) over a serial-port side-channel (well, it's more complicated than that, but anyway). I set up a listener for a failed connection status message which signal()s the server program to close all client connections (but not the bound dispatchers) and thereby force all clients to reconnect. It's a cheat and a cheesy hack, but it works. Other diagnostics: One test I intend to run in a couple of weeks (next opportunity) is to boot the CentOS 6.6 box with the older kernel, in order to find out whether the behavior change is in the kernel or in the libraries. Correct solutions: 1) Client port: The client should be connecting on a random, ephemeral port like a good client instead of on a fixed port, which I suspected. I don't know if this can be changed (due to a really dumb binary TCP driver). 2) Protocol change: The server never writes to the socket in the existing protocol, and can therefore never find out that the connection is dead. Writing to the socket would reveal this. But what happens if the server writes to the socket, and the client never reads? (We do, as it happens, have access to the client software, so the protocol can be fixed eventually. But I'm still curious as to the answer.) 3) Several people suggested using SO_REUSEADDR and/or an SO_LINGER of zero to drop the socket out of TIME_WAIT, but does the socket enter TIME_WAIT as soon as the client crashes? I didn't think so, but I may be wrong. 4) Several people suggested SO_KEEPALIVE, but those occur only after hours unless you change kernel parameters via procfs and/or sysctl, and when the client crashes, I need recovery right away, not hours down the road. Time here is literally worth a dollar per second, roughly. Anyway, thanks for the discusssion and helpful links. At one time I knew all this stuff, but it has been 20 years since I had to dig into the TCP protocol this deeply. -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Socket behavior change from 6.5 to 6.6
[I wish I knew how to get the mailing list to thread my replies properly in the archives; I subscribe to the daily digest, and replying to that doesn't do it.] Greg Lindahl wrote: On Thu, Jan 15, 2015 at 03:40:08PM -0300, Glenn Eychaner wrote: My only theory is that this has something to do with non-ephemeral ports and socket reuse, but I'm not sure what. If you want a quick detection that the link is dead, have the server occasionally send bytes to the dos box. You will get an immediate error if the dos box is up and knows that connection is kaput. What if I am sending bytes to the DOS box, but it never reads the socket? (Let us assume, for the sake of argument, that I can't change the DOS box software. In fact, I can, but it's more difficult than changing the Linux end.) Won't that either result in my detecting the socket as dead when it is not, or eventually overflowing the socket buffering? Given that the port numbers of the new connection are the same, I'm kind of surprised that the behavior changed from 6.5 to 6.6, but, I always use defensive programming (sending those extra bytes). I was super-surprised by the change, in that I fully tested the upgrade on my simulator system before deploying, and still got bit on deployment. Of course, the simulator doesn't have a real DOS box, just a simulation process that sends the images. [And, I also recently got bit by this http://www.macstadium.com/blog/osx-10-9-mavericks-bugs/ after upgrading some Macs. Sigh, network issues.] Alex from Germany wrote: Since you always use the same local port - maybe you need to set SO_REUSEADDR option. I assume I would have to set that on the client (DOS) side (the box which is using the same local port 1025 each time); setting it on the bound-listener socket on the Linux side doesn't seem like it would do anything to resolve the issue, based on my reading of SO_REUSEADDR on the net: http://www.unixguide.net/network/socketfaq/4.5.shtml http://stackoverflow.com/questions/14388706/ -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Socket behavior change from 6.5 to 6.6
I will try to explain this as best I can. I have two computers; one a Supermicro X10SAE running CentOS 6, the other a very old DOS box.[*] The DOS box runs a CCD camera, sending images via Ethernet to the X10SAE. Thus, the X10SAE runs a Python server on port 5700 (a socket which binds to 5700 and listens, and then accepts a connection from the DOS box; nothing fancy).[**] The DOS box connects to the server and sends images. This all works great, except: When the DOS box exits, crashes, or is rebooted, it fails to shut down the socket properly. Under CentOS 6.5, upon reboot, when the DOS box would attempt to reconnect, the original accepted server socket would (after a couple of connection attempts from the DOS box) see a 0-length recv and close, allowing the server to accept a new connection and resume receiving images. Under CentOS 6.6, the server never sees the 0-length recv. The DOS box flails away attempting to reconnect forever, and the server never seems to get any type of signal that the DOS box is attempting to reconnect. Possibly relevant facts: - The DOS box uses the same local port (1025) every time it tries to connect. It does not use a random ephemeral port. - The exact same code was tested on a CentOS 6.5 and 6.6 box, resulting in the described behavior. The boxes were identical clones except for the O/S upgrade. - The Python interpreter was not changed during the upgrade, because I run this code using my own 2.7.2 install. However, both glibc and the kernel were upgraded as part of the O/S upgrade. My only theory is that this has something to do with non-ephemeral ports and socket reuse, but I'm not sure what. It is entirely possible that some low-level socket option default has changed between 6.5 and 6.6, and I wouldn't know it. It is also possible that I have been relying on unsupported behavior this whole time, and that the current behavior is actually correct. Does anyone have any insight they can offer? [*] Hardware is not an issue; in fact, I have two identical systems, each of which has one X10SAE and three DOS boxes. But the problem can be boiled down to a single pair. [**] I'm actually using an asyncore.dispatcher to do the bind/listen, and then tossing the accept()ed socket into an asynchat. But I actually went ahead and put a trap on socket.recv() just to be sure that I'm not swallowing the 0-length recv by accident. -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] machine check exception
You have (AFAIK) provided no details as to which version of CentOS you are running nor of your hardware, but I'll try to help as I can. (In fact, it is unclear whether the MCE crashed your system or not!) I had a set of systems that occasionally logged MCEs (memory partity errors, in my case), and spent a month tearing into them. First, make sure that mcelog is installed on your system. If you are running 64-bit CentOS 6, you should be able to yum install mcelog. If you are running 32-bit CentOS 6 or CentOS 5, you'll have to download mcelog from the source (http://www.mcelog.org) and install it yourself, but if that is the case, let me know and I'll send further help. (I don't know about CentOS 7.) Second, make sure mcelogd is running at all times using system-config-services or chkconfig. Once you have done these two things, the next time you see an MCE, you should get an entry in /var/log/mcelog. This will tell you a LOT more about the MCE. Post the MCE here and/or Email it to me (I skim the digest and may miss a single post), and we can break it down further from there. [In my case, changing the memory had no effect on the MCEs, nor did any number of other suggested solutions; I eventually decided that since they were corrected memory parity errors, and thus non-fatal to processes or the system, I would ignore them. And as of the last kernel update, I don't see them any more, though I have not dug more deeply to see if there was some causal connection.] -G. On Oct 15, 2014, at 9:00 AM, centos-requ...@centos.org wrote: Unfortunately, No iLO Event Logs and IML Logs configured on the server. Can anybody suggest which tools on the server I can configure so next time server will have all the log records. Its really hard to prove to the peoples that the issue is at hardware level (When the Hardware vendor and Application Owners are from different companies ). -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Machine check events
m.roth writes: Is the system still under warranty? How 'bout the memory, if you've replaced it? You *should* replace it. It's not going to get better This is brand-new Kingston 1600MHz ECC memory on a workstation/server running at high altitude in a relatively open environment; I am loath to replace it based on a single correctable parity error every few days. Especially since both active computers are (thus far) seeing about the same error frequency (though it will take many more days or even weeks to determine that for certain; I haven't seen one in the last three days on either active computer), and memtest was run on these computers overnight (18+ hours) between build and deployment without apparent issue. [The computers were built in the states and then shipped 10,000 miles to the observatory location.] And the turnaround time from the observatory to the U.S. on servicing is no small matter. I have five of these computers (two active, one hot spare, one cold spare, one test system); if in the long run one proves to be a problem, i will deal with it at that time. If the memory is a bad batch, I'll need more proof. -G. On Nov 27, 2013, at 3:56 PM, Glenn Eychaner geycha...@mac.com wrote: And all that work was done to get this, output of a corrected memory parity error. I get about one of these per workstation per 3 days, more or less; is this a surprising number? (The workstation under the heaviest load gets more, while the idle spare gets none at all; no surprise there!) MCE 6 CPU 1 BANK 0 TIME 1385426237 Mon Nov 25 21:37:17 2013 MCG status: MCi status: Corrected error Error enabled MCA: Internal parity error STATUS 904f0005 MCGSTATUS 0 MCGCAP c09 APICID 2 SOCKETID 0 CPUID Vendor Intel Family 6 Model 60 -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Machine check events
On further, further, further toying, I now have mcelog running on my 32-bit CentOS 6 systems! I admit to doing it the dumb way: I grabbed the source from the git repository, compiled and installed it, and THEN discovered that the init.d file supplied with the source was not CentOS compatible, so I grabbed the x86-64 RPM, extracted the startup files, and copied them into place. The RPM was small enough to make this easy. What I SHOULD have done is to grab the source RPM, replace the source with the latest source, build and install the source RPM, and then repackage the RPMs again for future consumption. Maybe I will try that at a future date, but I don't really have time today. -G. On Nov 26, 2013, at 11:11 AM, Glenn Eychaner geycha...@mac.com wrote: On further, further investigation, it looks like according to the mcelog install guide at http://www.mcelog.org/installation.html, I could roll my own for 32-bit CentOS 6: For bad page offlining you will need a 2.6.33+ kernel or a 2.6.32 kernel with the soft offlining capability backported (like RHEL6 or SLES11-SP1) The kernel has to have CONFIG_X86_MCE enabled. For 32bit kernels you need at least a 2.6,30 kernel. The current kernel I am running is 2.6.32-358.23.2, but I can't tell whether it has CONFIG_X86_MCE enabled. How can I find this out? JD writes: yum info mcelog ... Description : mcelog is a daemon that collects and decodes Machine Check : Exception data on x86-64 machines. So not for 32-bit... On Nov 26, 2013, at 9:25 AM, Glenn Eychaner geycha...@mac.com wrote: Further investigation seems to indicate that these events should be handled by mcelog or mced. However, there is no /var/log/mcelog, nor do I have a mcelog or mced binary, nor does yum seem to contain anything related (based on yum whatprovides '*/mcelog' and similar queries). Thus, I still don't know what to do with these errors. Ignore them? I am running 32-bit CentOS 6.4 (legacy software reasons). On Nov 25, 2013, at 11:05 AM, Glenn Eychaner geycha...@mac.com wrote: On my new Haswell-based machines, I am occasionally seeing entries like the following in /var/log/messages: kernel: [Hardware Error]: Machine check events logged (I would not have even noticed them, except that they get flagged by logwatch.) These messages always occur alone, and don't seem to have a corresponding entry in any other log file in /var/log. How can I get more info about these messages? -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Machine check events
And all that work was done to get this, output of a corrected memory parity error. I get about one of these per workstation per 3 days, more or less; is this a surprising number? (The workstation under the heaviest load gets more, while the idle spare gets none at all; no surprise there!) MCE 6 CPU 1 BANK 0 TIME 1385426237 Mon Nov 25 21:37:17 2013 MCG status: MCi status: Corrected error Error enabled MCA: Internal parity error STATUS 904f0005 MCGSTATUS 0 MCGCAP c09 APICID 2 SOCKETID 0 CPUID Vendor Intel Family 6 Model 60 Anyway, -G. On Nov 27, 2013, at 3:32 PM, Glenn Eychaner geycha...@mac.com wrote: On further, further, further toying, I now have mcelog running on my 32-bit CentOS 6 systems! I admit to doing it the dumb way: I grabbed the source from the git repository, compiled and installed it, and THEN discovered that the init.d file supplied with the source was not CentOS compatible, so I grabbed the x86-64 RPM, extracted the startup files, and copied them into place. The RPM was small enough to make this easy. What I SHOULD have done is to grab the source RPM, replace the source with the latest source, build and install the source RPM, and then repackage the RPMs again for future consumption. Maybe I will try that at a future date, but I don't really have time today. -G. On Nov 26, 2013, at 11:11 AM, Glenn Eychaner geycha...@mac.com wrote: On further, further investigation, it looks like according to the mcelog install guide at http://www.mcelog.org/installation.html, I could roll my own for 32-bit CentOS 6: For bad page offlining you will need a 2.6.33+ kernel or a 2.6.32 kernel with the soft offlining capability backported (like RHEL6 or SLES11-SP1) The kernel has to have CONFIG_X86_MCE enabled. For 32bit kernels you need at least a 2.6,30 kernel. The current kernel I am running is 2.6.32-358.23.2, but I can't tell whether it has CONFIG_X86_MCE enabled. How can I find this out? JD writes: yum info mcelog ... Description : mcelog is a daemon that collects and decodes Machine Check : Exception data on x86-64 machines. So not for 32-bit... On Nov 26, 2013, at 9:25 AM, Glenn Eychaner geycha...@mac.com wrote: Further investigation seems to indicate that these events should be handled by mcelog or mced. However, there is no /var/log/mcelog, nor do I have a mcelog or mced binary, nor does yum seem to contain anything related (based on yum whatprovides '*/mcelog' and similar queries). Thus, I still don't know what to do with these errors. Ignore them? I am running 32-bit CentOS 6.4 (legacy software reasons). On Nov 25, 2013, at 11:05 AM, Glenn Eychaner geycha...@mac.com wrote: On my new Haswell-based machines, I am occasionally seeing entries like the following in /var/log/messages: kernel: [Hardware Error]: Machine check events logged (I would not have even noticed them, except that they get flagged by logwatch.) These messages always occur alone, and don't seem to have a corresponding entry in any other log file in /var/log. How can I get more info about these messages? -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Machine check events
Further investigation seems to indicate that these events should be handled by mcelog or mced. However, there is no /var/log/mcelog, nor do I have a mcelog or mced binary, nor does yum seem to contain anything related (based on yum whatprovides '*/mcelog' and similar queries). Thus, I still don't know what to do with these errors. Ignore them? I am running 32-bit CentOS 6.4 (legacy software reasons). -G. On Nov 25, 2013, at 11:05 AM, Glenn Eychaner geycha...@mac.com wrote: On my new Haswell-based machines, I am occasionally seeing entries like the following in /var/log/messages: kernel: [Hardware Error]: Machine check events logged (I would not have even noticed them, except that they get flagged by logwatch.) These messages always occur alone, and don't seem to have a corresponding entry in any other log file in /var/log. How can I get more info about these messages? -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Machine check events
On further, further investigation, it looks like according to the mcelog install guide at http://www.mcelog.org/installation.html, I could roll my own for 32-bit CentOS 6: For bad page offlining you will need a 2.6.33+ kernel or a 2.6.32 kernel with the soft offlining capability backported (like RHEL6 or SLES11-SP1) The kernel has to have CONFIG_X86_MCE enabled. For 32bit kernels you need at least a 2.6,30 kernel. The current kernel I am running is 2.6.32-358.23.2, but I can't tell whether it has CONFIG_X86_MCE enabled. How can I find this out? Thanks, -G. JD writes: yum info mcelog ... Description : mcelog is a daemon that collects and decodes Machine Check : Exception data on x86-64 machines. So not for 32-bit... On Nov 26, 2013, at 9:25 AM, Glenn Eychaner geycha...@mac.com wrote: Further investigation seems to indicate that these events should be handled by mcelog or mced. However, there is no /var/log/mcelog, nor do I have a mcelog or mced binary, nor does yum seem to contain anything related (based on yum whatprovides '*/mcelog' and similar queries). Thus, I still don't know what to do with these errors. Ignore them? I am running 32-bit CentOS 6.4 (legacy software reasons). On Nov 25, 2013, at 11:05 AM, Glenn Eychaner geycha...@mac.com wrote: On my new Haswell-based machines, I am occasionally seeing entries like the following in /var/log/messages: kernel: [Hardware Error]: Machine check events logged (I would not have even noticed them, except that they get flagged by logwatch.) These messages always occur alone, and don't seem to have a corresponding entry in any other log file in /var/log. How can I get more info about these messages? -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Machine check events
On my new Haswell-based machines, I am occasionally seeing entries like the following in /var/log/messages: kernel: [Hardware Error]: Machine check events logged (I would not have even noticed them, except that they get flagged by logwatch.) These messages always occur alone, and don't seem to have a corresponding entry in any other log file in /var/log. How can I get more info about these messages? Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] CentOS LiveCD on USB
I have been following these instructions: https://www.centos.org/forums/viewtopic.php?t=501 to put a bunch of utilities (Clonezilla, SystemRescue, CentOS netinstall/rescue, etc.) on a single USB key. It works great for everything (including Ubuntu Live) except the CentOS 6.4 LiveCD. (You can see my postings at the bottom of the forum.) When booting the LiveCD, I got: Kernel panic - not syncing: Attempted to kill init! Pid: 1, comm: init Not tainted 2.6.32-358.el6.i686 #1 After removing quiet and adding selinux=disabled, I got more information; the boot stalls after finding devices, and gives: No root device block:/dev/mapper/live-rw found dracut suggests adding rdshell, which I did. This was not helpful (I had no idea what to do in the dracut shell), but did notice that in the dracut shell /dev/ did NOT seem to contain my USB drive at /dev/sdb as I would expect. (One reason it seemeed not helpful) So: 1) I used VFAT rather than ext2/3/4. Do I have to use ext2/3/4? 2) Do I need to rebuild the initramfs file somewhere in the CentOS LiveCD directory? 3) Is this just a straight-up hardware incompatibility? The computer is a brand-new SuperMicro X10SAE Haswell system. Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] CentOS LiveCD on USB
I had already gotten rid of rghb. The grub2 entry on the key for booting the LiveCD reads: menuentry CentOS 6.4 Live { set root=(hd0,1) linux /CentOS-Live/isolinux/vmlinuz0 root=UUID=A352-6D7C ro liveimg nodiskmount nolvmmount selinux=disabled live_dir=/CentOS_Live/LiveOS initrd /CentOS-Live/isolinux/initrd0.img } The contents of the LiveCD appear in /CentOS_Live as one would expect. The boot fails right after a device scan (obvious by tens of lines listing ataN:, scsiN:, sd 0:0:0:0:, etc.) with the No root device error below. In the rdshell, /dev/sda shows up as the internal system hard drive rather than the USB key. The USB key does not show up as /dev/sdb nor any other device that I can find. Finally, I looked in /dev/mapper (duh); it contains /dev/mapper/control, but no /dev/mapper/live-rw. Sorry for any confusion, -G. m.roth wrote: Glenn Eychaner wrote: I have been following these instructions: https://www.centos.org/forums/viewtopic.php?t=501 to put a bunch of utilities (Clonezilla, SystemRescue, CentOS netinstall/rescue, etc.) on a single USB key. It works great for everything (including Ubuntu Live) except the CentOS 6.4 LiveCD. (You can see my postings at the bottom of the forum.) When booting the LiveCD, I got: Kernel panic - not syncing: Attempted to kill init! Pid: 1, comm: init Not tainted 2.6.32-358.el6.i686 #1 After removing quiet and adding selinux=disabled, I got more Get rid of rhgb, too. information; the boot stalls after finding devices, and gives: No root device block:/dev/mapper/live-rw found dracut suggests adding rdshell, which I did. This was not helpful (I had no idea what to do in the dracut shell), but did notice that in the dracut shell /dev/ did NOT seem to contain my USB drive at /dev/sdb as I would expect. When you boot from a USB key, it always shows as /dev/sda. Second, rdshell is a grub shell. Are you trying to boot from the USB? If so, I'd fix the grub menu on that, if it's on /dev/sda1 of the flash drive, to use /dev/sda2 for the root= -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] CentOS LiveCD on USB
On Nov 19, 2013, at 1:55 PM, Glenn Eychaner geycha...@mac.com wrote: I had already gotten rid of rghb. The grub2 entry on the key for booting the LiveCD reads: [...] linux /CentOS-Live/isolinux/vmlinuz0 root=UUID=A352-6D7C ro liveimg nodiskmount nolvmmount selinux=disabled live_dir=/CentOS_Live/LiveOS D'Oh! It was obvious right after I sent the message; underscore instead of dash in live_dir. Sigh. HOWEVER, even after correcting that, it STILL doesn't boot; same exact message as before. I have a suspicion that it's not finding the USB key during the device scan, given that I can't find the USB key in /dev in rdshell. Possibly a hardware incompatibility? (I haven't been able to test a LiveCD in the optical drive yet, but will do so now.) -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] CentOS LiveCD on USB
On Nov 19, 2013, at 2:07 PM, Glenn Eychaner geycha...@mac.com wrote: Possibly a hardware incompatibility? (I haven't been able to test a LiveCD in the optical drive yet, but will do so now.) The system boots a liveCD from the DVD drive just fine. It boots CentOS 6.4 from the hard disk. It boots everything BUT CentOS 6.4 LiveCD from the USB key. -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Problem with X11 application and Nouveau driver
Nvidia driver fixed the problem. Thank you El Repo! Where do I send you $10? (Seriously. This saved me so much time.) -G. Nicolas Thierry-Mieg wrote: If you just want an easy solution you could try the nvidia drivers from elrepo. Start with nvidia-detect to find out which version you need, as explained here: http://elrepo.org/tiki/kmod-nvidia On Nov 13, 2013, at 5:38 PM, Glenn Eychaner geycha...@mac.com wrote: I have finally received and am configuring my new workstations eith the NVS510 graphics cards, and have run into rather a problem. The X server seems to be loading the NOUVEAU driver properly (based on the contents of Xorg.0.log), but I have one X11 application that doesn't work correctly; it runs as though XSynchronized is always True, even though it's explicitly set to False in the code. In other words, its redraw behavior is god-awful; worse because it auto-redraws once per second for little apparent reason. And some of the windows draw with artifacts if they're covered and uncovered. Any ideas, anyone? I will gladly provide more info on request, but I'm not an expert X11 programmer (worse, this application uses a third party wrapper library). This is definitely new behavior on this new computer (the previous computers, also using the Nouveau driver with GeForce 7600 cards, did not seem to have this behavior). -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Problem with X11 application and Nouveau driver
I have finally received and am configuring my new workstations eith the NVS510 graphics cards, and have run into rather a problem. The X server seems to be loading the NOUVEAU driver properly (based on the contents of Xorg.0.log), but I have one X11 application that doesn't work correctly; it runs as though XSynchronized is always True, even though it's explicitly set to False in the code. In other words, its redraw behavior is god-awful; worse because it auto-redraws once per second for little apparent reason. And some of the windows draw with artifacts if they're covered and uncovered. Any ideas, anyone? I will gladly provide more info on request, but I'm not an expert X11 programmer (worse, this application uses a third party wrapper library). This is definitely new behavior on this new computer (the previous computers, also using the Nouveau driver with GeForce 7600 cards, did not seem to have this behavior). -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Cloning CentOS workstations
I manage a set of CentOS operations workstations which are all clones of each other (3 live and 1 spare kept powered down); each has a single drive with four partitions (/boot, /, /home, swap). I've already set up cron'd rsync jobs to copy the operations accounts between the workstations on a daily basis, so that when one fails, it is a simple, quick process to swap in the spare, restore the accounts from one of the others, and continue operations. This has been successfully tested in practice on more than one occasion. However, when I perform system updates (about once a month), I like to create a temporary clone of the system to an external drive before running the update, so that I can simply swap drives or clone back if something goes horribly wrong. I have been using CloneZilla to do this, but it can take a while since it blanks each partition before copying, and requires a system shutdown. Question 1: Would it be sufficient to simply use CloneZilla once to initialize the backup drive (or do it manually, but CloneZilla makes it easy-peasy), and then use rsync -aHx --delete (let me know if I missed an important rsync option) to update the clone partitions from then on? I am assuming that the MBR typically doesn't get rewritten during system updates, though /etc/grub.conf obviously does get changed. Suppose I want to store more than one workstation on a single drive (easy), and be able to boot into any of the stored configurations (hard). Here's what I thought of: 1) Create a small master partition which contains a bootloader (such as a CentOS rescue disk), and a single swap partition. 2) Create one partition set per workstation (/boot, /, /home, excluding swap). Obviously, these will all likely be logical, and each workstation must use unique labels for mounting partitions. 3) On the master partition, modify the bootloader menu to allow one to chainload the /boot partitions for each configuration. (This is the Voila! step that I haven't fully figured out.) Question 2: Is there a better way to do the above? How do I perform the Voila! step, i.e. what's the right chainload command for this? Also, the chainloaded partitions are logical; is this OK? I also have a single off-site NAS disk which contains clones of all the critical workstations on-site. Most of them are Macs, so I can use sparseimages on the NAS for the clones and get easy-peasy incremental clones. I also do this for the Linux box (backing it up incrementally to an HFS case-sensitive sparseimage via rsync), but it's (obviously) a bit of a kludge. Question 3: Is there a UNIX equivalent to the Mac sparseimage that I should be using for this? (tar -u can do it (duh), but then the backup file grows without bound.) Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Tk font problem with CentOS 6
I'm having an odd problem which I can't seem to find the answer to. I have recently upgraded from CentOS 5 to CentOS 6 (using a fresh install and migrate). However, I have a bunch of Tk widgets that use font names like 12x24 and 5x7, etc. Under CentOS 5 (Tk 8.4), this worked fine; however, under CentOS 6 (Tk 8.5), this does not seem to work properly; it does not find the fonts and reverts to a (pretty, but wrong) default font. I have verified using xlsfonts, xdpyfont, etc. that the fonts exist; for example, 12x24 is apparently now an alias for: -Sony-Fixed-Medium-R-Normal--24-170-100-100-C-120-ISO8859-1 and if I specify this full name in Tk, it works fine. What am I doing wrong here, or did the enhanced font support in Tk 8.5 subtly break fonts using an WxH designation? -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Intel 4600 Graphics (Haswell) in CentOS 6.4
Quick question that I haven't been able to find the answer to (and not for lack of trying, believe me): Is dual-monitor display for the new Intel HD Graphics 4600 (Haswell, e.g. Intel E3-1200v3 family processors) supported in CentOS 6.4? In particular, I'm looking at a SuperMicro X10SAE; SuperMicro has already replied that triple-display only works with a VGA-HDMI-DP combo (lame) and only in Windows (lame), but had no information as to dual-display (they reported to me that only tested single-head configurations [lame] before marking it as supported on their website). http://www.supermicro.com/support/resources/OS/C226.cfm I already found that the Intel opensource site only has releases for Fedora 19 and Ubuntu 13, and that X.org lists the latest release as 2011Q3 and the latest support as SandyBridge: http://01.org/linuxgraphics/downloads http://www.x.org/wiki/IntelGraphicsDriver/ But the latest version of xorg-x11-drv-intel in ElRepo Extras was uploaded in March of this year, and I haven't found whether the upstream vendor backported or sideported something into the latest distribution. [Yes, I'm STILL working on the workstation configuration. We're considering putting off triple-head support to cut costs, because the dreamy NVS510 cards are expensive. It's like a morass of quicksand sometimes.] Gracias y saludos, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] USB Audio sound card
On Aug 23, 2013, at 8:00 AM, centos-requ...@centos.org wrote: From: Fred Smith fre...@fcshome.stoneham.ma.us Subject: Re: [CentOS] USB Audio sound card On Thu, Aug 22, 2013 at 10:03:08PM -0400, Glenn Eychaner wrote: On Aug 22, 2013, at 3:11 PM, Glenn Eychaner geycha...@mac.com wrote: I apologize. I should have said here A quick search of the web (and the NewEgg comments) indicates that these devices generally work under *Linux*, but do they work in CentOS 6? good point. I should have known that's what you meant. however, it gives one hope. especially since many of those comments are a couple years old, it's given time for drivers to work their way into other distros--assuming the drivers were new at that time, and they may not have been. especially that first one you ask about is dirt cheap, so maybe the way to do it is to go buy one and try it. If I were in the U.S., I certainly would do that. As it is, I'm in Chile; if I can even find something similar here, it will likely be more expensive (I found one so far, but it's a high-end 5.1 model and costs USD$40), and ordering from the U.S. is a multiweek turnaround time. Hence, I decided to ask first and suffer the wrath of the list for asking the obvious. -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] USB Audio sound card
All- Ah, the saga of the 1U workstation continues. So, in all my work configuring the thing, I completely forgot about AUDIO; I only realized my mistake when I went on a cable-measuring expedition this morning. Unfortunately, none of the 1U servers I've been looking at come with audio outputs (there aren't even audio headers on the motherboard), and I've used the only availabnle slot for my fancy graphics card! Now, a lesser (or maybe smarter) individual would give up at this point, and go back to MiniITX or a 2U rackmount (if I could find a short-depth one). Nay! I say. What about USB Audio? I don't need 5.1 or 7.1 audio here; I'm plugging in a Dell monitor soundbar. A quick search of the web says that yes, these devices will work under CentOS and show up as /dev/dspX devices. So, do devices like these: SYBA SD-CM-UAUD USB Stereo Audio Adapter http://www.newegg.com/Product/Product.aspx?Item=N82E16812186035 Turtle Beach Audio Advantage Amigo II USB Interface Sound Card Headset Adapter http://www.newegg.com/Product/Product.aspx?Item=N82E16829118008 StarTech ICUSBAUDIO USB to Stereo Audio Adapter Converter http://www.newegg.com/Product/Product.aspx?Item=N82E16829128002 work under CentOS 6? Is there one that anyone can recommend? -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] USB Audio sound card
On Aug 22, 2013, at 3:11 PM, Glenn Eychaner geycha...@mac.com wrote: A quick search of the web says that yes, these devices will work under CentOS and show up as /dev/dspX devices. So, do devices like these: I apologize. I should have said here A quick search of the web (and the NewEgg comments) indicates that these devices generally work under *Linux*, but do they work in CentOS 6? [I have found in the past that Works in Ubuntu YY.MM, Works in Fedora N, do not always imply Works in CentOS/RHEL ; the driver support in CentOS/RHEL is sometimes more spartan than the cutting-edge distros. Of course, most of my experience is with CentOS 5; I only recently moved forward to CentOS 6 after extensive testing. What can I say? I'm as cutting edge as a dull butter knife.] SYBA SD-CM-UAUD USB Stereo Audio Adapter http://www.newegg.com/Product/Product.aspx?Item=N82E16812186035 Turtle Beach Audio Advantage Amigo II USB Interface Sound Card Headset Adapter http://www.newegg.com/Product/Product.aspx?Item=N82E16829118008 StarTech ICUSBAUDIO USB to Stereo Audio Adapter Converter http://www.newegg.com/Product/Product.aspx?Item=N82E16829128002 -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Slightly OT: PCIe x16 card in x8 slot
So, in the ongoing saga of the unusual 1U short-depth workstation, we have narrowed the field to two choices. Both entrants are configured with 16GB memory (4x4GB), two 2.5 drives (1x250GB SSD and 1x1TB HDD), and an NVIDIA NVS510 graphic card (quad display): 1) SuperMicro 5017R-MF, Xeon E5-2609 processor 2) SuperMicro 5017C-LF, Xeon E3-1220 processor (I wish SuperMicro had a list of their servers by chassis somewhere.) I have no preference between the solutions for right now, though I have a major concern with the second solution; the PCIe slot is only x8, and the NVS510 is a x16 card. The vendor assures me that a riser/adapter can be found to plug the card into the slot, and that it will work, but I am highly concerned about the performance. I know nothing about PCIe (I haven't built a system in 10 years, though I have read the PCIe Wikipedia entries and some guides); what kind of performance hit can I expect? Also, are there any solutions I have overlooked? Thanks again, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Triple- or Quad-display single-card graphics solutions
So, after some discussion of our new control workstations, we are iterating in on a solution; we are looking at a 1U short-depth SuperMicro SuperServer 5017R-MF with a graphics card in the PCI-Ex16 expansion slot. However, the display requirements have increased to 3 or more monitors for future expansion, so I was wondering whether anyone had any experience with triple- or quad-display single card solutions. Thus far, I have found two promising solutions: NVidia NVS 510 or 450 Matrox M-series M9138 or M9148 Both these solutions claim to have Linux support, but I was wondering if anyone had any experience with them in CentOS 6.4? And if there were any other solutions I had overlooked? Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Triple- or Quad-display single-card graphics solutions
Just found this thread http://lists.centos.org/pipermail/centos/2013-April/134212.html and Emailed the author for details. On Aug 20, 2013, at 2:52 PM, Glenn Eychaner geycha...@mac.com wrote: So, after some discussion of our new control workstations, we are iterating in on a solution; we are looking at a 1U short-depth SuperMicro SuperServer 5017R-MF with a graphics card in the PCI-Ex16 expansion slot. However, the display requirements have increased to 3 or more monitors for future expansion, so I was wondering whether anyone had any experience with triple- or quad-display single card solutions. Thus far, I have found two promising solutions: NVidia NVS 510 or 450 Matrox M-series M9138 or M9148 Both these solutions claim to have Linux support, but I was wondering if anyone had any experience with them in CentOS 6.4? And if there were any other solutions I had overlooked? -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Intel e1000e driver bug and 82574L controller
One of the more promising solutions I'm looking at for my dual-ethernet dual-monitor workstation contains an Intel 82574L Ethernet controller. I found a LOT of postings regarding a bug in the driver for this controller: http://www.doxer.org/learn-linux/resolved-intel-e1000e-driver-bug-on-82574l-ethernet-controller-causing-network-blipping/ https://bugzilla.redhat.com/show_bug.cgi?id=632650 but there isn't any clear indication as to whether the bug has been resolved in mainline CentOS 6.4 or not. Has this bug been resolved? -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Motherboard and chipset compatibility
So, having returned from a month's vacation, I'm back to work on attempting to build a set of small form factor CentOS compatible computers. I've really tried to do my homework, but this doesn't appear (at first glance) to be at all easy. It's not made easier by the fact that I have to get it right the first time (and I haven't built a PC in a decade); the time and money cost of shipping anything to and from my remote location in Chile means I can't afford to waste time buying and returning things. First question: does anyone have any experience with the Jetway NF9E-Q77 or ZOTAC Z77ITX-A-E motherboards? Having struck out on Intel Q77 or Z77-based SFF motherboards (the DQ77** series is completely out of stock everywhere, and the DZ77** series is ATX only), I have found a couple of Mini-ITX systems based on these two motherboards. Second question: Where can I get information about which Intel chipsets (Z77 vs Z87 vs Q77 vs C602 vs ...geez, there are a LOT of chipsets, as evidenced by http://www.supermicro.com/support/faqs/os.cfm) are supported by CentOS 6 / RHEL 6? I have not been able to find this information on either the Intel, RedHat, or CentOS web sites. Third (more general) question: My requirements are (I believe) modest: * 1U short-depth rackmount chassis OR Mini-ITX small-footprint chassis * Dual GbE network ports * Dual 1920x1200 monitor display * One SSD drive * 32-bit CentOS 6.4 compatible. It's the combination of the first, third, and fifth requirements that really seems to get me hung up. I've found plenty of 1U server systems (such as SuperMicro), but none of them support dual displays. (Some of them have a PCIe16x riser card that could conceivably accomodate a separate graphics card, assuming I could find one that fits; I have Emails in to various tech supports to inquire about this. I've found LOTS of 2U solutions, thanks, but only have 1U of available rack.) As far as Linux support goes, the RHEL Hardware List has thus far been pretty useless (much of the hardware on it is obsolete or discontinued), and most manufacturers' web sites have been equally useless. (One exception being ASUS, which has a Linux-compatibility list at http://www.asus.com/websites/global/aboutasus/OS/Linux.pdf SuperMicro has a very nice list referenced above, but none of their small form factor motherboards support dual displays AFAICT; I have found nothing useful at Intel's site.) Does anyone have any resources they'd like to point me to? Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Motherboard and chipset compatibility
m.roth at 5-cent.us wrote: Now, about what you're looking to build - you say that you want 1U, and mention rackspace: in my experience, rackmounts are a *lot* larger than a pizza box, so I'm a little confused at the requirements you're building for. The rack is already full; I only get that 1U of space by removing a spare part to another location, and unfortunately, I have a depth limit due to the power distribution module on the rack rear. These computers are replacing tower PCs that sit on the floor under a desk in a rather hostile environment, so I'd like to move them to either the desktop or the adjacent rack, but have limited space in either location (1U of short-depth rack or about room for a miniITX box on the desk). -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Motherboard and chipset compatibility
John R Pierce wrote: On 8/12/2013 9:14 AM, Glenn Eychaner wrote: * 1U short-depth rackmount chassis OR Mini-ITX small-footprint chassis * Dual 1920x1200 monitor display those two requirements together are unusual. most rackmount 1U systems are headless, except a basic VGA for initial configuration. dual display is generally found on a desktop system. I agree. In this case, the floor is not the best environment for the equipment, the adjacent rack has only 1U of short-depth rack space available, and the desktop is already crowded with keyboards and monitors. Since the reqirements are (relatively) modest (except those two), I was hoping to squeeze something in. Looks like I'm out of luck, and buying another full tower to hold a motherboard, a disk drive, and one expansion card. Sigh. -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Motherboard and chipset compatibility
Since the reqirements are (relatively) modest (except those two), I was hoping to squeeze something in. how about an ultrasmall form factor desktop, such as the Dell Optiplex 7010 USFF ? those have dual displayport outputs (requires $7 optional video output panel), and are 24x6.5x24cm I didn't even know that the Optiplex 7010 was CentOS compatible (though someone may have mentioned it in my previous thread); it is not on the RedHat Hardware List, not does Dell's web site go out of its way to mention it. Again, how does one find this kind of thing out? There has to be a better solution than 3 days of web searches, Emails to tech support, and forum posts. In addition, the USFF Optiplex seems to be limited to a Core i3 processor and a mere 2GB of memory, which while acceptable is not optimal (and worse than some other solutions I'm looking at). And for everyone suggesting KVMs, VMs, SSH, or other solutions...this is a telescope operations system, so none of those are really appropriate to the task, I'm afraid. I really want direct monitor/keyboard/mouse connections (and yes, I keep a hotspare warmed up at all times in case of a critical failure, and have had to use it on more than one occasion). And I'm sorry my postings don't seem to thread right in the archives. I subscribe to the Digest form orf the list and am compiling these replies using the web archives. Anyway, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] CentOS 6 SFF motherboard or complete system
I am trying to assemble or purchase a set of CentOS 6 compatible SFF workstations, and am finding it incredibly frustrating to do so. hardware.redhat.com is so slow as to be useless and provides almost no information about each of the 1,300 or so products listed in their database; clicking through them one at a time is incredibly frustrating (and about half of them are discontinued or out of stock when I actually go looking for them, like the Intel DQ series motherboards I was interested in). Vendor web sites are almost no use; they trumpet their Windows 8 compatibility all over the site, but finding information about Linux compatibility is next to impossible. My requirements aren't overwhelming; an i7 processor, four memeory slots preferred, dual 24 (1920x1200) monitor capability, and dual ethernet (or an expansion slot for a second Ethernet card). Anyone have any advice on how to attack this these days? I've been out of the hardware-purchase game on the Linux side for years, and most of my bookmarks no longer point anywhere useful, sadly. -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Centos 6.4: Possible bug in system-config-network-cmd
On May 21, 2013, at 4:52 PM, Glenn Eychaner geycha...@mac.com wrote: I'm having a puzzling problem with system-config-network-cmd in CentOS 6.4 This all works great, EXCEPT that if the machine is booted a fixed-IP profile, the the DHCP ifcfg file also winds up in /etc/sysconfig/network-scripts. So, if I have in profiles/dhcp/ifcfg-eth0_dhcp (with a hardlink in devices/, of course): After some fooling around, I figured out that the problem is with the default profile. When you switch profiles, it copies whatever devices are in the profile you switched to AND whatever is in the default profile into .../network-scripts/, and if you delete the default profile or remove all the network devices, it will *repopulate it for you* the next time you switch. And of course you can't just use the same generic name in all the profiles (ifcfg_eth0), because then it gets really confused since the device names in .../profiles/* have to match the devices in .../devices/. This is a CHANGE in behavior from CentOS 5, and whoever thought it was a good idea should be forced to use Windows ME for 30 days. -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] CentOS 6 and PHP
Still upgrading CentOS 5 to CentOS 6, and have run into the next issue: When I install httpd and php, everything works great, and the default-test ?php phpinfo(); ? works great. The problem is that I have a bunch of old HTML that seems to use ? alone (or worse, !--?), which worked in CentOS 5 but fails in CentOS 6. I can change all the HTML, but is there an easier way? Summary of previous issues: system-config-network-cmd leaving behind ifcfg files in network-scripts that it shouldn't: I still don't have an answer for this one; I work around it by naming all the scripts ifcfg-ethX_profile and cleaning any that don't match profile after running system-config-network-cmd. I suspect it has to do with either the contents of /etc/udev/rules.d/70-persistant-net.rules or the format of my ifcfg-ethX_* files, but haven't been able to resolve it. Probably will leave it as workaround works. UUIDs and boot drives: Didn't solve this one either, but again didn't try very hard because the recovery from a failed attempt is a royal pain. I decided that the actual UUIDs weren't important enough to me to matter, since I seldom boot with more than one workstation's drive connected at a time. I will probably go back to labels, since I can change those (apparently) without rendering the system unbootable. Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Changing disk UUID after cloning
So, I have a CentOS 6 system, and I want to make several clones of it. I'm using Clonezilla to clone the drives; that's no problem. But the drive UUIDs are driving me up the wall. After cloning, the two drives have the same UUID, but I'd like each clone to have different UUIDs so there's no possibility of a conflict when I am running diagnostics with two drives installed, etc. But when I change the UUID of the /boot or / partition (even if I update /etc/fstab), the system won't boot; it GRUBs OK (after I use recovery mode to rerun grub-install), but never gets to the 'Welcome to CentOS message. Do I need to rebless vmlinuz or initrd or initramfs in the /boot partition if I change the drive UUID? Or should I just ignore UUID and go back to using labels in /etc/fstab (which is what I did in CentOS 5)? Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Changing disk UUID after cloning
On May 22, 2013, at 4:14 PM, Reindl Harald h.rei...@thelounge.net wrote: Am 22.05.2013 21:58, schrieb Glenn Eychaner: So, I have a CentOS 6 system, and I want to make several clones of it. I'm using Clonezilla to clone the drives; that's no problem. But the drive UUIDs are driving me up the wall. After cloning, the two drives have the same UUID, but I'd like each clone to have different UUIDs so there's no possibility of a conflict when I am running diagnostics with two drives installed, etc. But when I change the UUID of the /boot or / partition (even if I update /etc/fstab), the system won't boot; it GRUBs OK (after I use recovery mode to rerun grub-install), but never gets to the 'Welcome to CentOS message. Do I need to rebless vmlinuz or initrd or initramfs in the /boot partition if I change the drive UUID? for the inital boot /etc/fstab is *irrelevant* logical thinking: if it can read it the partition is already mounted * at least GRUB config contains a line like root=UUID=b935b5db-0051-4f7f-83ac-6a6651fe0988 Not on my system; CentOS 6 uses grub 0.97, and my grub.conf file doesn't contain any UUIDs that I can find. * dracut / initramfs contains at least the UUID for /boot * did yiou try dracut -f after the changes? That's probably the problem; I will make another attempt in the morning, if I decide that I care. I may simply decide that I don't care if I have duplicated UUIDs between workstations, if it becomes too much trouble to fix. :-) -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Centos 6.4: Possible bug in system-config-network-cmd
I'm having a puzzling problem with system-config-network-cmd in CentOS 6.4. I have a workstation with a number of different grub boot configurations (a spare for a set of workstations, basically), each of which has a parameter MYHOST=hostname, and I am using system-config-network-cmd to set the boot configuration during the network process (using a small custom system service that runs just before network startup, reads the configuration name from /proc/cmdline and calls system-config-network-cmd -p configname). I have properly disabled NetworkManager, and have /etc/sysconfig/networking/devices and .../profiles set up correctly AFAIK (it was all copied from a CentOS 5 machine, and the hand links were maintained properly as needed; such a PITA that they got rid of the very nice GUI for this). This all works great, EXCEPT that if the machine is booted a fixed-IP profile, the the DHCP ifcfg file also winds up in /etc/sysconfig/network-scripts. So, if I have in profiles/dhcp/ifcfg-eth0_dhcp (with a hardlink in devices/, of course): TYPE=Ethernet DEVICE=eth0 HWADDR=MAC redacted BOOTPROTO=dhcp ONBOOT=yes USERCTL=no IPV6INIT=no PEERDNS=yes and in profiles/fixed/ifcfg-eth0_fixed GATEWAY=x.y.z.1 TYPE=Ethernet DEVICE=eth0 HWADDR=MAC redacted BOOTPROTO=none NETMASK=255.255.255.0 IPADDR=x.y.z.n ONBOOT=yes USERCTL=no IPV6INIT=no PEERDNS=yes If I boot into fixed, I find that ifcfg-eth0_dhcp is also in network-scripts, and it tries to activate this interface, even though this interface is NOT in profiles/fixed in any way! This worked great in CentOS 5, so I think I know what I'm doing here? For now, I am fixing the issue by running a find on network-scripts to remove inappropriate files after running systme-config-network-cmd, but that is complete cheese, of course. Is there something missing from ifcfg-eth0_dhcp that is confusing the system-config-network-cmd script? Is there any documentation on this that's helpful? And is there simply a better way to do this that I've missed? Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Centos 6.4: Possible bug in system-config-network-cmd
m.roth: Hmmm... have you looked at /etc/udev/rules.d/70-persistant-net.rules? # PCI device 0x8086:0x104b (e1000e) SUBSYSTEM==net, ACTION==add, DRIVERS==?*, ATTR{address}==00:16:xx:xx:xx:xx, ATTR{type}==1, KERNEL==eth*, NAME=eth0 # PCI device 0x10b7:0x9200 (3c59x) SUBSYSTEM==net, ACTION==add, DRIVERS==?*, ATTR{address}==00:04:xx:xx;xx:xx, ATTR{type}==1, KERNEL==eth*, NAME=eth1 Not sure there's anything relevant there... -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Configuring printers in CentOS 5
Could someone please explain to me how to best configure printers in CentOS 5? I've been trying to configure a new printer, which is served by a Mac Mini: If I open a web browser at localhost:631, or system-configure-printers and I configure the new printer as an IPP printer, it winds up in a list of Remote printers, and once it winds up there I can't seem to delete the printer or change the settings at all. If I try using the system-config-printers interface, it's all greyed out; if I try using CUPS, it actually tries to connect to CUPS on the Mini! In order to delete it, I have to manually revert the /etc/cups files and restart the cups printing system in system-config-services. I just don't understand printer configuration on Linux. On the Macs, it's just plug and go. Why does CentOS have to make it so confusing? -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] projects.centos.org down again
Garry Dale wrote: There is an open bug report from 2009-09-21 with a similar summary [1]. Since bug 3858 was never closed, I've updated the notes. Per updates to bug tracker, the projects.centos.org site is back online. Should bug 3858 [1] now be closed, or is it acting as a placeholder for events such as this? Just curious... Well, projects.centos.org was up briefly over the weekend, but appears (from my end) to be down again, same symptoms; HTTP connections just hang. Next time I'll download the LiveCD instructions I'm looking for to a static file! -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Projects.centos.org down?
I have been trying to get to the CentOS LiveCD site at projects.centos.org the last couple of days, but have been unable to reach it. Is it down, and is there any info on when it might be back up? Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] CentOS, Firefox, and Java Plugin
On Sep 22, 2010, at 11:22 PM, Glenn Eychaner wrote: The latest updates to CentOS 5.5 seem to have broken the Java plugin, and have defeated any and all attempts to get it working again. I'm running CentOS 5.5 (32-bit) and Firefox 3.6.9 (installed from the CentOS repository); I've tried BOTH the openJDK plugin available through the Argeo repositories, and installing Java 1.6.0 directly from Sun/Oracle and creating the plugin soft link in /usr/lib/mozilla/plugins. Neither works at all. Thanks to everyone for their help. It turns out that I had two problems: 1) The page that you get redirected to by the Firefox plugin finder links to these (incorrect) install instructions: http://java.com/en/download/help/linux_install.xml#rpm The correct install instructions can be found at Oracle's website: http://www.oracle.com/technetwork/java/javase/install-linux-rpm-137089.html 2) I was using the Argeo-Plus plugin (32-bit), which appears to be broken. Mathieu, if you get it working again, I'd be happy to use it, but if not, I understand. I was aware that the location had changed (serendipitously), and have the latest version (I've since rolled everything to the Oracle/Sun plugin). Thanks all, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] CentOS, Firefox, and Java Plugin
The latest updates to CentOS 5.5 seem to have broken the Java plugin, and have defeated any and all attempts to get it working again. I'm running CentOS 5.5 (32-bit) and Firefox 3.6.9 (installed from the CentOS repository); I've tried BOTH the openJDK plugin available through the Argeo repositories, and installing Java 1.6.0 directly from Sun/Oracle and creating the plugin soft link in /usr/lib/mozilla/plugins. Neither works at all. This was working a while ago, but it broke and I didn't notice. Thanks for any help you can provide, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] System beeps in kernel 2.6.18-194
On Aug 26, 2010, at 12:00 PM, centos-requ...@centos.org wrote: ?Is there someplace can I find *detailed* release notes on the differences between -164 and -194 kernels to help in looking for the problem, pinning it down, and submitting a patch (and/or building my own kernel), or should I just download the SRPMS and dig in? You can find kernel changelog diffs here (maintaind by Alan Bartlett): http://www.centos.toracat.org/ajb/kernel-clog-diff/ Is there any place that I can find RPM or SRPM packages for the kernels between 18-164 and 18-194 that are listed in this kernel log? It would really help narrow down the problem if I could just build each kernel version and test it; then at least I'd only have one set of differences to go through rather than 30. [Not a kernel expert, but willing to give it a solid go!] Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] System beeps in kernel 2.6.18-194
On Aug 31, 2010, at 9:36 AM, Glenn Eychaner wrote: On Aug 26, 2010, at 12:00 PM, centos-requ...@centos.org wrote: ?Is there someplace can I find *detailed* release notes on the differences between -164 and -194 kernels to help in looking for the problem, pinning it down, and submitting a patch (and/or building my own kernel), or should I just download the SRPMS and dig in? You can find kernel changelog diffs here (maintaind by Alan Bartlett): http://www.centos.toracat.org/ajb/kernel-clog-diff/ Is there any place that I can find RPM or SRPM packages for the kernels between 18-164 and 18-194 that are listed in this kernel log? It would really help narrow down the problem if I could just build each kernel version and test it; then at least I'd only have one set of differences to go through rather than 30. Having read the kernel log diff list, and searched it for items related to sound, I'm really suspicious of the following listed change: * Mon Dec 21 2009 Jarod Wilson ja...@redhat.com [2.6.18-183.el5] [...] - [sound] alsa hda driver update for rhel5.5 (Jaroslav Kysela) [525390] How do I go about backing out this change (reverting the alsa hda drivers in the -194 kernel to the -164 kernel versions) for testing? -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] System beeps in kernel 2.6.18-194
So, just today I noticed a problem with kernel 2.6.18-194 (CentOS 5.5) on several Intel DP965LT systems; the system beeps (such as terminal beeps) are no longer passed through to the external speakers. This is a problem because in our situation the boxes are distant from their monitor/keyboard, the system speaker on this motherboard is extremely weak, and there are no system speaker header pins on the motherboard. The problem goes away if I revert the system to 2.6.18-164 with no other changes. I looked through the list archives and searched the web for other people who have encountered this, but it's pretty specific (and hard to search for system beep!) Is there someplace can I find *detailed* release notes on the differences between -164 and -194 kernels to help in looking for the problem, pinning it down, and submitting a patch (and/or building my own kernel), or should I just download the SRPMS and dig in? Thanks, -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] System beeps in kernel 2.6.18-194
Akemi Yagi amyagi at gmail.com wrote: On Wed, Aug 25, 2010 at 2:39 PM, Glenn Eychaner geychaner at mac.com wrote: So, just today I noticed a problem with kernel 2.6.18-194 (CentOS 5.5) on several Intel DP965LT systems; the system beeps (such as terminal beeps) are no longer passed through to the external speakers. This is a problem because in our situation the boxes are distant from their monitor/keyboard, the system speaker on this motherboard is extremely weak, and there are no system speaker header pins on the motherboard. The problem goes away if I revert the system to 2.6.18-164 with no other changes. I looked through the list archives and searched the web for other people who have encountered this, but it's pretty specific (and hard to search for system beep!) Does it have Nvidia controllers? If so, it may be related to: http://bugs.centos.org/view.php?id=4335 Nope. It has a PCI NVidia graphics card, but the tech specs for the board (page 12) show Intel chipsets (as expected). http://downloadmirror.intel.com/15049/eng/DP965LT_TechProdSpec.pdf Besides, I tried adding the enable_msi=0 to that line of modprobe.conf, and it didn't make any difference. Is there someplace can I find *detailed* release notes on the differences between -164 and -194 kernels to help in looking for the problem, pinning it down, and submitting a patch (and/or building my own kernel), or should I just download the SRPMS and dig in? You can find kernel changelog diffs here (maintaind by Alan Bartlett): http://www.centos.toracat.org/ajb/kernel-clog-diff/ Wow. Considering I need to look at everything from 164-15 to 194-3, that's a lot of heavy reading and searching. -G. -- Glenn Eychaner (geycha...@lco.cl) Telescope Systems Programmer, Las Campanas Observatory ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos