Latest posts of: entropy

Show Posts
Pages: [1]

1	Linux Stuff / Kernel / Re: sheeva/dream 3.2.13 new kernel available	on: April 11, 2012, 10:45:19 PM
Quote from: entropy on April 11, 2012, 10:38:34 PM tar: usr: Cannot open: No such file or directory # readlink /lib; ls -l lib usr/lib lrwxrwxrwx 1 root root 7 Dec 31 1969 lib -> usr/lib #

2	Linux Stuff / Kernel / Re: sheeva/dream 3.2.13 new kernel available	on: April 11, 2012, 10:38:34 PM
Quote from: cbxbiker61 on March 26, 2012, 12:42:41 AM ok, it's done. BTW on my systemd systems I decided NOT to make /lib, /bin and /sbin symlinks to /usr/? for this exact reason. It's much safer to make the items inside the directory links when necessary. In any case the scripts should handle this problem effectively now. Not quite. I get this error: Extracting modules tar: usr: Cannot open: No such file or directory tar: Error is not recoverable: exiting now

3	Linux Stuff / Kernel / Re: sheeva/dream 3.2.13 new kernel available	on: March 25, 2012, 09:10:24 AM
You might want to update your README-PLUG-UPDATE.sh script to check for /lib being linked to /usr/lib (like in Fedora 17+). Currently the script will overwrite the /lib link and make any system that relies on this unbootable. The safest thing would probably be to check if /lib is a symlink, follow it, and untar there.

4	Linux Stuff / General Linux questions / Re: Guruplug USB harddisk, journal i/o errors	on: January 19, 2011, 05:44:15 PM
Quote from: Nestor on January 14, 2011, 07:30:49 AM But within a few seconds the kernel reassigns a new block device. Plugging it in for the first time the block device might be /dev/sdc1, but when the journal I/O errors appear it disconnects and reconnects and appears under a new block device name, for example /dev/sdd1. Then I remount under the new name, but soon I get the same errors again and it disconnects/reconnects again and choses /dev/sde1. This keeps incrementing. Any chance you're using the plug to power the drive? I had a very similar symptom but it was caused by the fan going out. The fan would drag down with resistance and slow down. That would drag down the power supply voltage and everything on the USB bus would reset. If you're powering the drive from the plug, you might be dragging down the power supply voltage.

5	Linux Stuff / Kernel / Re: 2.6.36.2 new kernel available	on: January 04, 2011, 11:12:48 PM
Quote from: cbxbiker61 on January 01, 2011, 12:48:53 PM No. There won't be until the iscsi target project releases a newer version than 1.4.20.2 that supports kernels > 2.6.35. Since 2.6.35.x is a long term support kernel it should be good to go for everything. I can't think of any critical features .36 adds for sheevaplug. Interesting.... Before I saw your reply, I was messing around with scsi-target-utils and got my plug working as an iSCSI target... Please don't tell my plug Perhaps what you meant to say was that the kernel was complete but the iscsi target project userland is not compatible with the 2.6.36?

6	Linux Stuff / Kernel / Re: 2.6.36.2 new kernel available	on: January 01, 2011, 11:45:32 AM
Quote from: cbxbiker61 on December 11, 2010, 05:11:36 PM 2.6.36.2 is now available. Is everything required for iSCSI target enabled again in the 2.6.36.x kernels?

7	Linux Stuff / Kernel / Re: 2.6.36.2 new kernel available	on: January 01, 2011, 11:04:39 AM
Quote from: cbxbiker61 on December 29, 2010, 03:00:30 PM HPET only exists on x86. ONDEMAND is only valid for cpu's that scale frequency. Which the sheevaplug can't do. Since powertop was written by Intel for Intel, quite a few of it's recommendations are x86 specific. Someone could submit a patch to the powertop maintainers that would turn off invalid recommendations for the sheevaplug. Makes sense. Thanks for checking.

8	Linux Stuff / Kernel / Re: 2.6.36.2 new kernel available	on: December 29, 2010, 12:58:10 PM
I've been running powertop on my plug (kernel 2.6.36). Here are two suggestions from powertop: Suggestion: Enable the CONFIG_CPU_FREQ_GOV_ONDEMAND kernel configuration option. The 'ondemand' CPU speed governor will minimize the CPU power usage while Suggestion: Enable the CONFIG_HPET_TIMER kernel configuration option. Without HPET support the kernel needs to wake up every 20 milliseconds for some housekeeping tasks. http://cateee.net/lkddb/web-lkddb/CPU_FREQ_GOV_ONDEMAND.html This one might save some heat by being able to throttle down a CPU. http://cateee.net/lkddb/web-lkddb/HPET_TIMER.html Not sure this one would be useful at all (does the guruplug/sheevaplug have a HPET timer? can it even be enabled for a ARM processor?)

9	Hardware and U-Boot firmware / U-Boot stuff / Re: ESIA - does it work with GuruPlug?	on: October 12, 2010, 04:07:19 PM
It worked fine with my guruplug plus.

10	General Category / General Discussion / Re: New Guruplug is noisy, but does it work?	on: October 06, 2010, 08:10:02 PM
I just received the PLUS version. Install a new uboot, kernel, and OS. It will boot from SDHC (which is why I got the PLUS). I'm running single gig-e, no bluetooth, no wireless, and it's running plenty cool. Mine has been at a load of about 1.2 for the past few days with no signs of trouble. It's going to be a file/media server.

11	Linux Stuff / Kernel / Re: 2.6.35.7 new kernel available	on: October 02, 2010, 02:45:08 PM
Can you add: CONFIG_TASKSTATS, CONFIG_TASK_DELAY_ACCT, and CONFIG_TASK_IO_ACCOUNTING? # iotop Could not run iotop as some of the requirements are not met: - Python >= 2.5 for AF_NETLINK support: Found - Linux >= 2.6.20 with I/O accounting support (CONFIG_TASKSTATS, CONFIG_TASK_DELAY_ACCT, CONFIG_TASK_IO_ACCOUNTING): Not found #

12	Linux Stuff / Kernel / Re: 2.6.35.7 new kernel available	on: September 30, 2010, 04:45:32 PM
I'd like to recommend for the future you turn timestamps back off. Most people don't want or need them and they just add extra boot noise. These options are easily accessed by those who do need/want them through the standard kernel command line parameters: http://www.kernel.org/pub/linux/kernel/people/gregkh/lkn/lkn_pdf/ch09.pdf (see earlyprintk and time parameters). Unfortunately, I do not see a way to disable them via command line if they are already enabled.

Pages: [1]