dimanche 1 mars 2015

CentOS Software RAID 1 Issue [on hold]


I have configured the CentOS Software RAID 1 with two hard disk of 250 GB. I checked the status of the hard disk in raid.Then I tried to boot the hard disk in separate but the hard disk did not boot in the OS and it shows Kernel Panic Error.I have configured the Software RAID 1 by using the following link


http://ift.tt/1AYmnYF


Then, I made one of the hard disk as degraded and added new hard disk for sync the data. The syncing of data done successfully but when I tried to boot the synced harddisk alone also means the same kernel error coming.Please follow this link where I configured as such in my raid device http://ift.tt/1AL9anp


What do I need to do so that the RAID 1 device can be booted from a single hard disk alone?



Converting .vhd file to .raw - File is too small and does not boot


I have a Hyper-v .vhd file I am trying to convert to .raw for Proxmox however every time I convert the size is less than half of the actual disk space used and when I import it doesn't boot.


I download the .vhd directly off my Windows machine and run the following command to convert


qemu-img convert -f vpc -O raw Rami.vhd Rami.raw


Then import


dd if=Rami.raw of=/dev/bigdata/vm-102-disk-1


I also tried with qcow2 but it doesn't work also(file size too small and doesn't boot). Does the disk size matter? This is a 300GB .vhd file and I have done the exact same thing with other .vhd files less than 50GB with no problems.


What am I missing?



how to choose appropriate linux hardware - poor graphics performance


I can't believe I am asking this question here myself. Apologies if this is not the right place for this question.


I've been using linux since 1996, SuSE version 4.1...


And yet I am still not able to put together a well performing machine.


I bought a simple box a few months ago. I am a software developer. I don't need high end graphics nor RAID nor stuff like that.


My main goal is to have a fast responsive machine. I installed arch linux on it, and i3 as desktop environment - really it should be quite fast.


And yet...the machine responsiveness is very low, especially graphics lag A LOT. I know I didn't opt for a high end graphics engine - but shouldn't such a modern machine be enough for lag-free youtube and other video watching? Lag-free firefox browsing?


Maybe I just misconfigured my machine, and that's what my question is about. Some hardware info follows below.


As you can see, I have a AMD E-350 processor with 8GB-RAM - I should think this would be quite enough for a standard linux installation...


What I also did is, as the MoBo comes with a GPU/processor bundle, I thought that maybe upgrading just the graphics card may have some impact - nope. I bought a AMD HD5450 card with 1GB RAM (!!!) - and still I can't enjoy a fluent video on youtube, nor a sufficient experience web browsing...(maybe it's even better to have the original GPU than the added card I bought?). Of course I know that modern browsers are pretty demanding, but I should think such a machine should be able to handle it?


I read the posts on the arch linux forum about optimizing video, and AMD cards are not very well reviewed really...


There are infinite posts about how to choose your distro, there are hardware compatibility lists - but are there any tutorials or guides which actually help setting up a fast and responsive machine?



  • Is my machine maybe badly configured?

  • Does it just have a bad hardware set up? Should I maybe replace one or the other piece to cheaply get satisfactory performance?

  • Am I just asking too much from my cheap hardware?


I am considering replacing the whole MoBo (I know it was cheap and I shouldn't expect too much from it, but I thought I should be able to tweak it...)


(lshw):



*-memory
description: System memory
physical id: 0
size: 7967MiB
*-cpu
product: AMD E-350 Processor
vendor: Advanced Micro Devices [AMD]
physical id: 1
bus info: cpu@0
size: 800MHz
capacity: 800MHz
width: 64 bits
capabilities: fpu fpu_exception wp vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp x86-64 constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni monitor ssse3 cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch ibs skinit wdt arat hw_pstate npt lbrv svm_lock nrip_save pausefilter vmmcall cpufreq
*-pci:0
description: Host bridge
product: Family 14h Processor Root Complex
vendor: Advanced Micro Devices, Inc. [AMD]
physical id: 100
bus info: pci@0000:00:00.0
version: 00
width: 32 bits
clock: 66MHz
configuration: latency=32
*-pci:0
description: PCI bridge
product: Family 14h Processor Root Port
vendor: Advanced Micro Devices, Inc. [AMD]
physical id: 4
bus info: pci@0000:00:04.0
version: 00
width: 32 bits
clock: 33MHz
capabilities: pci normal_decode bus_master cap_list
configuration: driver=pcieport
resources: irq:16 ioport:e000(size=4096) memory:fea00000-feafffff ioport:c0000000(size=268435456)
*-display
description: VGA compatible controller
product: Cedar [Radeon HD 5000/6000/7350/8350 Series]
vendor: Advanced Micro Devices, Inc. [AMD/ATI]
physical id: 0
bus info: pci@0000:01:00.0
version: 00
width: 64 bits
clock: 33MHz
capabilities: vga_controller bus_master cap_list rom
configuration: driver=radeon latency=0
resources: irq:41 memory:c0000000-cfffffff memory:fea20000-fea3ffff ioport:e000(size=256) memory:fea00000-fea1ffff


How to diagnose potential JFFS2 flash problems?


I have a large fleet of embedded ARM systems using the JFFS2 filesystem. Are there any particular commands or log file entries I can query to gauge whether they are having any corruption or other problems?



Can you do a btrfs snapshot of a live system?


I am planning on giving btrfs a try soon, and what I want to know, and can't seem to find online is this:


Can you do a btrfs snapshot on a running system if the snapshot is not just of some small directory but of the entire system (/var, /usr, /bin, /sbin, ... everything), or do you need to boot into a live CD to do a btrfs snapshot while the system is offline?


If it is technically possible to do a snapshot of a running system, I assume you still need to shut the system down to use that snapshot. Would booting a snapshot (subvolume) of a system that was running at the time the snapshot was taken be very similar to a hard boot, since software is not aware when a snapshot is about to take place, so programs might be writing something to /var at the time, and next thing they know, the program is starting up again with data half written because the system restarted?



Source code location for debugging on Debian?


From Debian's DebugPackage, I know the debug information file for an executable goes in /usr/lib/debug when creating two part executables (i.e., stripping the executable of its symbols and placing them in a separate file).


However, the symbol file will refer to source files, and I don't see where the source files are supposed to be placed. On Red Hat/Fedora, I know they are located in /usr/src/debug (according to Fedora's Packaging:Debuginfo).


Where do I put source files for debugging on Debian?




A concrete example is Crypto++. I have a patch for its GNUMakefile that adds the following:



IS_DEBIAN = $(shell uname -a 2>&1 | $(EGREP) -i -c "debian|ubuntu|mint")
...

# http://ift.tt/1ECVxs0
ifeq ($(IS_DEBIAN),1)
DEBUG_SYM_DIR ?= /usr/lib/debug/cryptopp
DEBUG_SRC_DIR ?= /usr/src/debug/cryptopp
endif
...


And then there's a symbol recipe that looks like so:



symbol symbols:
$(MKDIR) -p $(DEBUG_SYM_DIR) $(DEBUG_SRC_DIR)
-objcopy --only-keep-debug cryptest.exe cryptest.exe.debug
-objcopy --only-keep-debug libcryptopp.so libcryptopp.so.debug
-strip --strip-debug --strip-unneeded cryptest.exe
-strip --strip-debug --strip-unneeded libcryptopp.so
-$(CP) cryptest.exe.debug $(DEBUG_SYM_DIR)/
-$(CP) libcryptopp.so.debug $(DEBUG_SYM_DIR)/
-objcopy --add-gnu-debuglink=$(DEBUG_SYM_DIR)/cryptest.exe.debug cryptest.exe
-objcopy --add-gnu-debuglink=$(DEBUG_SYM_DIR)/libcryptopp.so.debug libcryptopp.so
-$(CP) *.h *.cpp $(DEBUG_SRC_DIR)/


So the workflow is:



cd cryptopp
make static dynamic test
sudo make symbols
sudo make install


Related, but not relevant (in case someone wants to comment): I can't add symbols as a dependency to another recipe or just make symbols because of a bug in objcopy (I believe its a bug - see Binutil Bug 18064 - objcopy, add-gnu-debuglink and "cannot fill debug link section").



samedi 28 février 2015

Why is Firefox so slow over SSH?


I try to launch Firefox over SSH, using



ssh -X user@hostname


and then



firefox -no-remote


but it's very very slow.


How can I fix this? Is it a connection problem?