- luv-main - example.com

Centos 6 rpm yum issues
by Andy Dean 03 May '13

03 May '13

Hi Has anyone come across a centos 6 64bit system where rpm and yum just return a prompt after entering them. Thanks Sent from my Note 2

1 0

SMART self-test failed but drive still healthy
by James Harper 02 May '13

02 May '13

I have just had two drives failed in a server today. One is mostly part of a RAID0 set (which is in turn part of a DRBD, so we're still good) and a small partition that is part of a RAID1, which hasn't been failed (errors are about 1.3TB along a 2TB disk). The other is one I was testing, it wasn't particularly new and doesn't really matter. Both drives have logged read errors under Linux kernel, both report drive is healthy status (SMART overall-health self-assessment test result: PASSED), and both say "Completed: read failure" almost immediately when I do a SMART self test (short test or long). I don't really have any trouble with the fact that two drives have failed, but I'm really surprised that SMART still reports that the drive is good when it is clearly not... what's with that? James

4 8

Ruxmon Intro Security Talk on Hooking / Tracing Binaries on Linux
by Andrew Worsley 01 May '13

01 May '13

Thought this might be of interest to Linux (often security talks are windows issues related) and other people needing to analyse strange binaries under Linux. Andrew ------- Ruxmon Melbourne is on this Friday. As usual, we will adjourn to the Oxford Scholar for a meal and a bit of a meet and greet. Ruxmon is a free monthly community security event organised and run by the Ruxmon team. We are currently looking for some new Ruxmon speakers. Please email me if you would like to speak on anything security related for between 10-30 minutes in front a small and non-intimidating audience. For more information please visit our website: http://www.ruxmon.com Presentations Introductory level hooking in Linux - Ryan Platt This talk will walk through methods of tracking execution of Linux binaries, including an example of a simple Linux Kernel Module hooking the system call table and usermode hooking by patching binaries on disk.Introductory level hooking in Linux. Introducing Daisho: an open source multi-protocol network tap - Dominic Spill Every communication technology should have a widely available network tap, preferably operating as close to the physical layer as possible. If we can't tap the comms, assessing their security is a much harder task. As we try to squeeze more bandwidth out of our links the problem gets even tougher. Our solution to this problem is project Daisho; an open source hardware and software project to build a device that can monitor high speed communication links and pass all of the data back to a host system for analysis. Daiso will include a modular, high bandwidth design that can be extended to monitor future technologies. The project will also produce the first open source USB 3.0 FPGA core, bringing high speed data transfer to any projects that build on the open platform. As a proof of concept at this early stage, Dominic will demonstrate monitoring of a low bandwidth RS-232 connection using our first round of hardware and discuss the challenges involved with the high speed targets that we will take on later this year. http://www.ruxmon.com/Melbourne/ Details Date: Friday, 26th April Time: 6:00PM Location: RMIT University, City Campus https://my.rmit.edu.au/portal/page/portal/RMITPortal/campusmaps?dsize=max Room 080.09.012 (Building 80, Level 9, Room 12) RMIT Building 80 entrance is off Swanston Street (just past Swanston and A'Beckett St) and next door to the Oxford Scholar Hotel. Please take the lift to Level 9 and make your way to Room 12.

3 3

mysql very slow inserts
by James Harper 27 Apr '13

27 Apr '13

I've posted the question on the bacula-users list too but everyone is asleep there I think. My Bacula installation has suddenly slowed to a crawl and I've tracked it down (I think) to temporary table inserts. I've switched Bacula to use attribute spooling so it's fairly obvious that the backup job runs really fast (to the limit of USB2 connected backup medium) but then at the end of the job when it inserts the records into the database it almost stops. The backup job took about 2 minutes to copy a few GB of data but the database operation is still running a few hours later - it is making progress just slowly. It seems that Bacula has created a temporary table called batch and is inserting records into that, and then at the end I think it inserts the records into the main job table. The system load is ~1.15, and mysqld is highest on the list of processes but is only 1-2%, which I assume means that the cpu is being blocked somewhere due to IO (?). Disk write performance inside the VM where Bacula and mysql runs is around 90MB/second, which is more than sufficient to insert a few hundred thousand rows. iostat while all this is going on shows pretty much nothing happening. I think this has happened before but then came good before I got time to look at it properly. Can anyone please recommend how I can figure out where it's all going wrong? I've used the general log in mysql to determine that it's only the INSERT operations running, and iostat and strace to look for anything obviously loaded, but have come up empty. Thanks James

5 13

Re: [luv-talk] New Computer System
by Russell Coker 27 Apr '13

27 Apr '13

On Sat, 27 Apr 2013, David Zuccaro <david.zuccaro(a)optusnet.com.au> wrote: > Also I don't have a video card, I'm using an onboard graphics > controller: > > 00:02.0 VGA compatible controller: Intel Corporation 82945G/GZ > Integrated Graphics Controller (rev 02) The Intel video chipsets have a history of being well supported for the common functions. So while 3G etc will usually perform badly the basic functions including scaling of videos will usually perform well. Every time I've compared Intel video on a motherboard to a AGP or PCIe video card that performs better for 3G I've found the Intel hardware to work better for playing movies. > This is the motherboard: > > http://www.gigabyte.com/products/product-page.aspx?pid=2304#sp > > I put in an extra 2GB stick so now its memory capacity has been maxed > out. http://en.wikipedia.org/wiki/List_of_Intel_chipsets According to the above Wikipedia page you are limited to 2G or 4G of RAM with that motherboard as you have a 945GZ chipset. > Using mplayer seems to give much better performance than totem. As a general rule if there are multiple programs to perform a task then testing another if the first one doesn't work well enough is a good strategy. Video players are very complex and there's lots of different options, you really don't want to know about all the work that goes in to optimising them. But it's complex enough that there are lots of ways for a player to not be optimised ideally for some systems. > > When iceweasel performs poorly how much swap is in use and is the hard > > drive light on a lot? > > No. This is not a swap issue. No swap is in use just complex (buggy > resource intensive, badly put together web pages. I think I will make do > with what I have at this stage. I will definitely need a bigger HD and > as you say swapping to a new HD is going to be a PITA but it needs to be > done. top also gives memory usage. Last time I tested Chromium seemed to perform better with slow Javascript. But Chromium does use more RAM. As you've upgraded your RAM you should try Chromium and see how it goes for the slow sites. Even if you don't like Chromium it's still good to know where the problems are. > In summary I think I will do some upgrading before I buy a new system; > and look into getting a graphics card. Thanks everyone for your > comments. http://www.graysonline.com/lot/0001-611545/computers-and-it-equipment/dell- optiplex-780-small-form As an aside, the fastest CPU supported by your motherboard is the type of CPU in systems that some people put out as e-waste. Buying a faster system at auction isn't going to cost much. The above auction has a system with a CPU that's considerably faster than the fastest that is supported on your current motherboard and the current price is $100 with less than a day to go. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

1 0

Speaking IPP by hand
by trentbuck＠gmail.com 26 Apr '13

26 Apr '13

I have a printer that's being strange. Since the printer itself runs ipp, I'm trying to send it jobs directly with lp/lpstat/loptions/cancel &c, without talking to cupsd. I get some responses, but nothing very helpful. Are the cups client utilities supposed to be able to talk generic IPP, or do they assume the far end is cups? Since IPP is based on HTTP, I also tried just using netcat, but RFC2910 §13.1 seems to be sying that it has a bunch of byte-encoded stuff, rather than just VERB LOCATION PROTO/VERSION. Google and IRC haven't helped, so my next step I guess is to actually read the RFCs instead of just skimming them. $ ssh root@printserver grep DeviceURI /etc/cups/printers.conf DeviceURI ipp://mfd [....] $ ping -c1 mfd PING mfd.cyber.com.au (203.7.155.91) 56(84) bytes of data. 64 bytes from MFD.cyber.com.au (203.7.155.91): icmp_seq=1 ttl=254 time=0.723 ms --- mfd.cyber.com.au ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 0.723/0.723/0.723/0.000 ms This saves adding -h every time. Also I found that lpstat -t -h foo will STILL talk to the default printserver (not foo) for some of its queries -- WTF? $ cat ~/.cups/client.conf ServerName mfd $ lpstat -t scheduler is running no system default destination lpstat: server-error-operation-not-supported lpstat: server-error-operation-not-supported lpstat: server-error-operation-not-supported lpstat: server-error-operation-not-supported $ date | lp -d canthappen request id is canthappen-498 (0 file(s)) $ date | lp -d foo request id is foo-500 (0 file(s)) $ date | lp -d '' request id is -502 (0 file(s)) ...printer still hasn't printed anything. $ lpstat -t scheduler is running no system default destination lpstat: server-error-operation-not-supported lpstat: server-error-operation-not-supported lpstat: server-error-operation-not-supported lpstat: server-error-operation-not-supported MFD speaks PS (I think), so try that as well as text/plain. Doesn't help. $ date | a2ps -o- | lp -d '' # the mfd speaks PS (I think), so try that. [stdin (plain): 1 page on 1 sheet] [Total: 1 page on 1 sheet] sent to the standard output request id is -504 (0 file(s)) $ nc mfd ipp GET / HTTP/1.1 400 Bad Request Server: gSOAP/2.7 Content-Length: 0 Connection: close $ nc mfd ipp Print-Job 1 IPP/1.1 HTTP/1.1 400 Bad Request Server: gSOAP/2.7 Content-Length: 0 Connection: close $

2 2

maybe broken SD card
by Russell Coker 26 Apr '13

26 Apr '13

Recently the micro SD card in my Samsung Galaxy S3 stopped working. As I wasn't sure whether the card or the phone was broken I tried connecting it to my Thinkpad T61 running Debian/Unstable. It gave the following log entries when I inserted the card (in a holder to make it fit the full size SD socket in the Thinkpad). Is this proof that the micro SD card is broken? I will of course try other cards in the phone (which needs more storage) and try the card in another phone (it probably has data I want to keep). But I lack enough hardware to be sure that there isn't some strange partial compatability issue. Apr 25 20:03:30 linux kernel: [ 599.446191] mmc0: error -110 whilst initialising SD card Apr 25 20:03:30 linux kernel: [ 599.449766] sdhci-pci 0000:15:00.2: Will use DMA mode even though HW doesn't fully claim to support it. Apr 25 20:03:30 linux kernel: [ 599.514728] sdhci-pci 0000:15:00.2: Will use DMA mode even though HW doesn't fully claim to support it. Apr 25 20:03:30 linux kernel: [ 599.584550] sdhci-pci 0000:15:00.2: Will use DMA mode even though HW doesn't fully claim to support it. Apr 25 20:03:30 linux kernel: [ 599.659497] sdhci-pci 0000:15:00.2: Will use DMA mode even though HW doesn't fully claim to support it. As an aside, where can I get a cheap micro SD card? -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

5 5

Re: [luv-talk] New Computer System
by Russell Coker 24 Apr '13

24 Apr '13

On Mon, 22 Apr 2013, Craig Sanders <cas(a)taz.net.au> wrote: > if your m/b takes DDR2 RAM then it' still possible that upgrading RAM > alone may be enough, but it's a bigger gamble....if it doesn't work, > you'll just be throwing the money away unless you have a use for a 2nd > computer after you buy a new one. DDR2 is pretty much obsolete, and > more expensive than newer DDR3 - 4GB of DDR2 costs $56. Of course if they are going to keep using the old system then upgrading the RAM won't be such a waste. Having a second PC is a really good idea for anyone who's serious about a computer hobby and it could also be given to a friend or relative. > Alternatively, if you do go for a new system with an i7 CPU, you may > find that the built-in Intel GPU is good enough - they're low-end if > you're a gamer, but more than adequate for video playing, and they have > good open source drivers. My experience of Intel video cards is that they are quite good for playing video with the free Linux drivers. So far the only time I've found a built-in Intel video controller to be inadequate for my use (which is occasional video playing and other tasks that aren't particularly challenging - I'm not a serious gamer) is when I got a monitor with a resolution higher than FullHD. > The FX-8350 @ $209 is $100 cheaper than the i7-3770, and a decent AM3 > motherboard is also about $50 to $100 cheaper (e.g. Sabertooth Z77 for > i7 @ $244 vs Sabertooth 990FX for AMD CPUs at $197) http://www.graysonline.com When I'm buying new hardware I visit Grays Online. You can get entire systems for less than the cost of the CPUs you list. Sure they won't be as fast, but I'd rather buy a new system every year or so than wait years for a big upgrade. > as Russell suggested, a RAID-1 array is better/safer. but if you don't > need the extra terabyte of storage, 2x2TB drives is a lot cheaper than > the 2x3TB drives he suggested. http://www.tecs.com.au/shop/storage/desktop-sata-hard-drives.html?limit=30 At the above there's a Seagate 2TB disk for $119 and a Seagate 3TB disk for $139. Not a lot cheaper, only $20 for the extra TB. There's a WD Green 2TB disk for $99 and a WD Green 3TB disk for $159, so if you want the WD Green series then there's a significant price difference. But generally you won't want WD Green for a RAID array due to the head parking issues. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

3 2

Re: [luv-talk] New Computer System
by Russell Coker 24 Apr '13

24 Apr '13

On Mon, 22 Apr 2013, "Trent W. Buck" <trentbuck(a)gmail.com> wrote: > I assume you'd have a non-negligible amount of zlib CPU churn when > paging out to zram -- is that noticable? I suppose it is, but only > when you have pegged BOTH CPU and RAM. Using zram is a trade-off of CPU performance vs disk IO performance. As CPU performance has been increasing steadily at an exponential rate for the last 20 years and disk performance has been increasing at a slow linear rate over that period such a trade-off becomes increasingly beneficial. Pegging both CPU and RAM is something that's difficult to do given that running out of RAM results in processes being blocked on disk IO (regular file IO that's not in cache or swap). A quick test on an Intel T7500 CPU (which is far from the fastest CPU available today and wasn't even the fastest laptop CPU when I bought it) shows that gzip -9 on a 10MB file takes half a second of CPU time. While this is quite a bit of time (more than it would take to do a direct contiguous write of the same data) when you count random seeks for small access it's going to average a much higher rate than HDD access. Rates lower than 3MB/s are often seen on modern disks in real-world use due to seek overhead. Whether zram is better than SSD is another issue. With the Intel SSD I've tried the contiguous read and write speeds are quite a bit lower than that of hard disks and while the overall performance is a significant improvement over hard disks it's still not nearly as good as I'd hoped for. If SSD was as good as some people claim then zram might not offer much benefit. But with the 120G Intel SSDs I've tried there is plenty of scope for zram to boost performance. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

3 2

zram
by Russell Coker 24 Apr '13

24 Apr '13

There has been some discussion on luv-talk (which should have been on luv- main) about zram, a system that uses a compressed RAM disk for swap on Linux. I've enabled zram on one of my systems. I ran "free" on all the systems I run and found that only one had any significant swap use, that's a workstation with 8G of RAM and 512M of swap in use. I didn't bother enabling it on any other systems. I added "zram" to /etc/modules (modules that are loaded on boot on Debian). I created /etc/modprobe.d/zram.conf with the following contents: options zram num_devices=2 This gives two devices and it seems that you are expected to have one zram device for each CPU core. I'm not sure if this is really required but it doesn't seem to do any harm. To enable zram on boot I added the following to /etc/rc.local: echo 1073741824 > /sys/block/zram0/disksize echo 1073741824 > /sys/block/zram1/disksize mkswap /dev/zram0 mkswap /dev/zram1 swapon -p 100 /dev/zram0 swapon -p 100 /dev/zram1 The above sets each of the two devices to 1G of storage and enables them as swap space. I haven't yet rebooted the system to make sure that there are no errors in the above, but as it's what I types on the command-line it should work. Also I haven't tested zram extensively. Finally I'm running the Debian/Experimental 3.8 kernel. I don't know how it would work on other kernels. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

1 0