Search Results

Search found 227 results on 10 pages for 'zac thompson'.

Page 4/10 | < Previous Page | 1 2 3 4 5 6 7 8 9 10  | Next Page >

  • How can a single disk in a hardware SATA RAID-10 array bring the entire array to a screeching halt?

    - by Stu Thompson
    Prelude: I'm a code-monkey that's increasingly taken on SysAdmin duties for my small company. My code is our product, and increasingly we provide the same app as SaaS. About 18 months ago I moved our servers from a premium hosting centric vendor to a barebones rack pusher in a tier IV data center. (Literally across the street.) This ment doing much more ourselves--things like networking, storage and monitoring. As part the big move, to replace our leased direct attached storage from the hosting company, I built a 9TB two-node NAS based on SuperMicro chassises, 3ware RAID cards, Ubuntu 10.04, two dozen SATA disks, DRBD and . It's all lovingly documented in three blog posts: Building up & testing a new 9TB SATA RAID10 NFSv4 NAS: Part I, Part II and Part III. We also setup a Cacit monitoring system. Recently we've been adding more and more data points, like SMART values. I could not have done all this without the awesome boffins at ServerFault. It's been a fun and educational experience. My boss is happy (we saved bucket loads of $$$), our customers are happy (storage costs are down), I'm happy (fun, fun, fun). Until yesterday. Outage & Recovery: Some time after lunch we started getting reports of sluggish performance from our application, an on-demand streaming media CMS. About the same time our Cacti monitoring system sent a blizzard of emails. One of the more telling alerts was a graph of iostat await. Performance became so degraded that Pingdom began sending "server down" notifications. The overall load was moderate, there was not traffic spike. After logging onto the application servers, NFS clients of the NAS, I confirmed that just about everything was experiencing highly intermittent and insanely long IO wait times. And once I hopped onto the primary NAS node itself, the same delays were evident when trying to navigate the problem array's file system. Time to fail over, that went well. Within 20 minuts everything was confirmed to be back up and running perfectly. Post-Mortem: After any and all system failures I perform a post-mortem to determine the cause of the failure. First thing I did was ssh back into the box and start reviewing logs. It was offline, completely. Time for a trip to the data center. Hardware reset, backup an and running. In /var/syslog I found this scary looking entry: Nov 15 06:49:44 umbilo smartd[2827]: Device: /dev/twa0 [3ware_disk_00], 6 Currently unreadable (pending) sectors Nov 15 06:49:44 umbilo smartd[2827]: Device: /dev/twa0 [3ware_disk_07], SMART Prefailure Attribute: 1 Raw_Read_Error_Rate changed from 171 to 170 Nov 15 06:49:45 umbilo smartd[2827]: Device: /dev/twa0 [3ware_disk_10], 16 Currently unreadable (pending) sectors Nov 15 06:49:45 umbilo smartd[2827]: Device: /dev/twa0 [3ware_disk_10], 4 Offline uncorrectable sectors Nov 15 06:49:45 umbilo smartd[2827]: Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error Nov 15 06:49:45 umbilo smartd[2827]: # 1 Short offline Completed: read failure 90% 6576 3421766910 Nov 15 06:49:45 umbilo smartd[2827]: # 2 Short offline Completed: read failure 90% 6087 3421766910 Nov 15 06:49:45 umbilo smartd[2827]: # 3 Short offline Completed: read failure 10% 5901 656821791 Nov 15 06:49:45 umbilo smartd[2827]: # 4 Short offline Completed: read failure 90% 5818 651637856 Nov 15 06:49:45 umbilo smartd[2827]: So I went to check the Cacti graphs for the disks in the array. Here we see that, yes, disk 7 is slipping away just like syslog says it is. But we also see that disk 8's SMART Read Erros are fluctuating. There are no messages about disk 8 in syslog. More interesting is that the fluctuating values for disk 8 directly correlate to the high IO wait times! My interpretation is that: Disk 8 is experiencing an odd hardware fault that results in intermittent long operation times. Somehow this fault condition on the disk is locking up the entire array Maybe there is a more accurate or correct description, but the net result has been that the one disk is impacting the performance of the whole array. The Question(s) How can a single disk in a hardware SATA RAID-10 array bring the entire array to a screeching halt? Am I being naïve to think that the RAID card should have dealt with this? How can I prevent a single misbehaving disk from impacting the entire array? Am I missing something?

    Read the article

  • Logitech Performance MX Mouse Jumps on OS X Lion (10.7.4)

    - by Adam Thompson
    I have a Logitech MX Revolution wireless mouse that I am trying to use with OS X Lion. Everything is working except for one problem... there is a small, but quite noticeable, jump when the mouse cursor is moved. The problem is mostly prevalent when dragging and dropping files or trying to highlight items. It makes performing any task with the mouse accurately next to impossible. I did quite a bit of looking and found that all kinds of people have had mouse issues with OS X. I've tried all of the following with absolutely no success: Using the official drivers from Logitech (these performed worse than the default mouse drivers in OS X) Using SteerMouse as a third party mouse driver. This worked ever so slightly better than the default driver, but still suffered quite frequently from the skipping problem Cleaning the sensor on the mouse and ensuring it's not the result of the surface that it's being used on. Tested the mouse on a Windows machine. The mouse worked absolutely flawlessly on the other machine. Changed the channel that my wireless router operates on by the off chance my problems were the result of interference. This also had no effect. I can't think of anything else that could possibly interfere with the mouse. I'm am out of ideas on what to try, so I would really appreciate if anyone has any suggestions. I should also mention that an old wired mouse I had laying around worked just fine when I plugged it in. This really isn't the best solution, however, as I really prefer the MX Revolution.

    Read the article

  • How do I make a privileged port non-privileged in Redhat 5?

    - by Jason Thompson
    So I have a RedHat 5 box that I'm wanting to run an application that I wrote that implements SLP. SLP uses port 427 for answering service queries. My understanding is that ports below 1024 are "privileged" and thus cannot be bound to by anyone that's not root. I cannot run this application as root as it is launched via tomcat. One creative solution I really like was simply writing an iptables rule to route the privileged port to a non-privileged. In my proof of concept tests, this works wonderfully. Unfortunately, it would be greatly (and understandably) desired by the powers if my application did not require screwing around with iptables upon installation. So I heard a rumor and cannot find anything to verify this that there was some sort of command or parameter that could be set to make any port I want be non-privileged. Is this true? If so, how is this done? Thanks!

    Read the article

  • What is the value/cost of enabling "spread spectrum clocking" on my hard drives?

    - by Stu Thompson
    I'm building up a biggish NAS box (10x WD RE4 2TB SATA RAID10) and ran into some problems. During the course of my research, debugging, investigations, etc, I discovered a jumper on the physical drives labeled "spread spectrum clocking". After some googling about this on teh internets, it seems to be a feature that some suggest (without reference or explanation) enabling in 'a storage configuration' that makes the drive less sussesptable to EMI. But why? I've got three core questions: Why is this feature not enabled by default? What are the actual benefits? Are there any costs?

    Read the article

  • gitweb refusing to blame

    - by Slipp D. Thompson
    I'm attempting to get gitweb (git 1.8.4.2, via git instaweb) in a project dir on my Debian server to offer blame views. In my /etc/gitweb.conf: … # default logo, favicon, etc. settings $feature{'blame'}{'default'} = [1]; $feature{'pickaxe'}{'default'} = [1]; $feature{'snapshot'}{'default'} = ['tgz', 'txz', 'zip']; $feature{'highlight'}{'default'} = [1]; $feature{'pathinfo'}{'default'} = [1]; In my global config file: [gitweb] blame = true snapshot = tgz, txz, zip patches = 256 avatar = gravatar [instaweb] local = false httpd = apache2 -f port = 4321 In my project's .git/config file: [gitweb] blame = true And yet, when I try to load a git blame view (via hand-modifying the URL to http://myserversip:4321/?p=.git;a=blame;f=Tests/InchCoordProxyTests.m;h=b4b2…;hb=53b4, since blame action links don't show up): Doing a quick search for “Blame view not allowed” in the gitweb.cgi source reveals plainly that the gitweb_check_feature('blame') conditional is failing. What am I doing wrong? Or, is there a way to verbosely print out why gitweb is doing what it's doing (e.g. which config files were read, which settings were loaded from each file, etc.)?

    Read the article

  • Is the decision to use SNI or IP based SSL made during cert purchase or cert installation?

    - by Neil Thompson
    It's time to renew an SSL cert - but the website will soon be moving from a dedicated machine with a fixed IP to a cloud based host behind a load balancer. When I renew or re-purchase my ssl cert do I make the decision about whether it should be an SNI / IP based SSL Cert at the point of purchase - or is a cert a cert and it's all about where and how it's installed? I'm hoping the renewed cert can continue to be IP based for now, and in a few months when the website (and it's domain ofc) moves to the cloud I can re-use the cert in 'SNI mode'

    Read the article

  • Open Directory authenticated bind succeeds, but creates incomplete record

    - by Jay Thompson
    I have about a dozen Macs running 10.6.7 or 10.6.8, which are all failing to bind properly to my new 10.7.4 Server OD. I can bind them just fine via Directory Utility or dsconfigldap, and it reports success. However, when I look at the record, it is failing to write the MAC address. Even if I manually update the record with the MAC address, MCX doesn't do anything and clients can't log in to OD accounts. All of the affected clients have hundreds of lines in the /Library/Logs/DirectoryService.error.log like so: 2012-09-15 22:23:18 EDT - T[0x00007FFF70292CC0] - GetMACAddress returned 0x *** bad control string *** 8x I do know that all of these clients were previously managed with the Guest computer account, and I also know that they were all imaged with a DeployStudio image when they were purchased. I've tried dscacheutil -flushcache, but after that I'm drawing a blank. Google has a few hits, but nothing very helpful. Re-imaging would be ideal but probably isn't going to happen. Anyone come across this before?

    Read the article

  • New to server admin, Diagnosing Memory and CPU issues on DV

    - by G Thompson
    Sorry for my ignorance and lack of knowledge. I'm a PHP/Front-end developer just now venturing into very minor server management/diagnostics. I have a Media Temple DV account. I have 2 sites that run a PHP script through a subscription service to an API. Basically API hits site with said script. Script runs, gathers data from api, saves data to SQL database. I noticed that these sites seemed to causing memory overages on my server (not sure why). So I temporarily disabled them. The memory overage alerts stopped but my CPU still sits really high, like at 115% and above. I'm trying to diagnos this with tutorials and resources but just can't seem to find a solution. I'll attach screenshots(screenshots are without the PHP scripts I assume are responsible for the memory issues) I'm assuming are important to the diagnosis, but if anyone can point me in the right direction to start A. figuring out if and why the PHP script may be causing memory overages and B. Why my CPU is always over 100%. Thanks guys! Links to screen shots... can't post with low points. http://i.stack.imgur.com/A64k4.png http://i.stack.imgur.com/qm1rV.png

    Read the article

  • Is it possible with Google searches to ban any and all results from a domain? [closed]

    - by Stu Thompson
    Is it possible to configure Google somehow to permanently ban search results from domains that I know 100% are never, ever going to make me happy? Something cookie/session based maybe? E.g. I want to ban (permanently, forever and always) results from experts-exchange.com. Every time I click results that take me to their page I just want to scream. Update! Google has released a Chrome Extension to allow users to block individual site from Google search results! Personal Blocklist (by Google). (Since this question has been closed, I cannot answer it.)

    Read the article

  • La vulnérabilité de Java déjà exploitée depuis des serveurs russes, Oracle reste sur sa position

    Mise à jour du 15/04/10 La vulnérabilité de Java déjà exploitée Depuis des serveurs russes, Oracle reste sur sa position La faille de Java récemment mise à jour par un ingénieur de Google (lire ci-avant) serait déjà exploitée. Roger Thompson, chef chercheur chez AVG, a repéré des attaques depuis des serveurs russes utilisés par des sites qui ciblent le grand public (comme Songlyrics.com, qui propose les paroles de chansons de Lady Gaga, Rihanna, etc.). En arrivant sur ce site, un iFrame malicieux camouflé dans une publicité redirige l'utilisateur (sans que celui-ci ne s'en aperçoive) vers un serveur hébergeant l'exploit....

    Read the article

  • Echo styles into the mashup of a google map and wordpress custom fields

    - by zac
    Using wordpress, I am pulling in a custom fields from specific posts to fill in the content for a google generated map. I am using this code var point = new GLatLng(48.5139,-123.150531); var marker = createMarker(point,"Lime Kiln State Park", '<?php $post_id = 182; $my_post = get_post($post_id); $title = $my_post->post_title; $snip = get_post_meta($post_id, 'mapExcerpt', true); echo $title; echo $snip; ?>') map.addOverlay(marker); I am trying to echo css style blocks but this causes a javascript error var point = new GLatLng(48.5139,-123.150531); var marker = createMarker(point,"Lime Kiln State Park", '<?php $post_id = 182; $my_post = get_post($post_id); $title = $my_post->post_title; $snip = get_post_meta($post_id, 'mapExcerpt', true); echo "<div class='theTitle'>"; echo $title; echo "</div>"; echo $snip; ?>') map.addOverlay(marker); I get the error missing ) after argument list and the output is var point = new GLatLng(48.5139,-123.150531); var marker = createMarker(point,"Lime Kiln State Park", '<div class='theTitle'>Site Title</div>Site excerpt') map.addOverlay(marker); Can someone please show me a more elegant (working) solution for this?

    Read the article

  • Data Formatters temporarily unavailable

    - by Zac Altman
    Im trying to use Date Formatters (UIDateFormatter), but I keep getting this error: Program received signal: “EXC_BAD_ACCESS”. Data Formatters temporarily unavailable, will re-try after a 'continue'. (Unknown error loading shared library "/Developer/usr/lib/libXcodeDebuggerSupport.dylib")

    Read the article

  • Redirect help fixing broken backslash link

    - by zac
    I just accidentally launched a campaign that is driving traffic to index.html/ (backslash is breaking it)... I am not sure I can access the .htaccess and was wondering if I were to create a page so that people directed to this path would see soemthing what would the name of that page be? Since there is no name just an extra slash I am not sure how to handle this.

    Read the article

  • UISegmentedControl makes UITableView slow/lag?

    - by Zac Altman
    So I have a nicely working UITableView consisting of 3 rows (each including and image, and a varying number of text fields). Now the 4th row has a UISegmentedControl. As soon as I added it, the UITableView lags/jumps/skips. When I take it away again, everything is smooth. How can I add the UISegmentedControl and still have smooth scrolling?

    Read the article

  • URL filtering for UIWebView on the iPhone

    - by Zac Altman
    Can someone please shed some light on how I would get this to work: http://www.icab.de/blog/2009/08/18/url-filtering-with-uiwebview-on-the-iphone/ I tried making the "FilteredWebCache.h" and "FilteredWebCache.m" files in my project, but it said that "FilterManager.h" did not exist. What am I meant to do with those files? This I put in viewDidLoad: NSArray *paths = NSSearchPathForDirectoriesInDomains(NSDocumentDirectory, NSUserDomainMask, YES); NSString *docDir = [paths objectAtIndex:0]; NSString *path = docDir; // the path to the cache file NSUInteger discCapacity = 10*1024*1024; NSUInteger memoryCapacity = 512*1024; FilteredWebCache *cache = [[FilteredWebCache alloc] initWithMemoryCapacity:memoryCapacity diskCapacity:discCapacity diskPath:path]; [NSURLCache setSharedURLCache:cache]; [cache release];

    Read the article

  • iPhone long plist

    - by Zac Altman
    I have some data i want to add in to my app...about 650 categories (includes a name + id number), each with an average of 85 items (each with a name/id number). Will the iPhone support such a large plist? I want to first display the categories in a UITableView, when a category is selected I want to display all of the associated items. Having such a large plist, im not sure if the iPhone will lag when loading the items. At over 51,000 lines it seems like...it might.

    Read the article

  • Subversive Connector Discovery error

    - by Zac
    Trying to install Subversive for the first time. Successfully installed Subversive "Team Provider" via the Available Sites dialog in Help Install New Software. Added a site that points to the Helios Update Site. Restarted Eclipse and tried to open the SVN Repository Browser and it automatically launched a Subversive Connector Discovery dialog, presenting me with 6 or 7 connectors to choose from. I've tried every single one, and they all fail, stating that an "error occured" and that I need to check the "error log" (Eclipse error log?) for details. I then exited the dialog and noticed that I can now view all the SVN-related perspective panes. (1) Was it alright for me to cancel that Connector Discovery process, or will I need that to create/edit repositories locally? If I need it, what was going wrong and does anyone have any ideas for how to get it working? (2) I'm trying to undertand the SVN Repositories feature for creating a new repository. How do I point it to my local SVN server instance? The default is to query me for a URL that I assume I would point to one of those "publicly available" repos.... which I don't want! Thanks for any and all help here!

    Read the article

  • C++ resize a docked Qt QDockWidget programmatically?

    - by Zac
    I've just started working on a new C++/Qt project. It's going to be an MDI-based IDE with docked widgets for things like the file tree, object browser, compiler output, etc. One thing is bugging me so far though: I can't figure out how to programmatically make a QDockWidget smaller. For example, this snippet creates my bottom dock window, "Build Information": m_compilerOutput = new QTextEdit; m_compilerOutput->setReadOnly(true); dock = new QDockWidget(tr("Build Information"), this); dock->setWidget(m_compilerOutput); addDockWidget(Qt::BottomDockWidgetArea, dock); When launched, my program looks like this: http://yfrog.com/6ldreamidep (bear in mind the early stage of development) However, I want it to appear like this: http://yfrog.com/20dreamide2p I can't seem to get this to happen. The Qt Reference on QDockWidget says this: Custom size hints, minimum and maximum sizes and size policies should be implemented in the child widget. QDockWidget will respect them, adjusting its own constraints to include the frame and title. Size constraints should not be set on the QDockWidget itself, because they change depending on whether it is docked Now, this suggests that one method of going about doing this would be to sub-class QTextEdit and override the sizeHint() method. However, I would prefer not to do this just for that purpose, nor have I tried it to find that to be a working solution. I have tried calling dock-resize(m_compilerOutput-width(), m_compilerOutput-minimumHeight()), calling m_compilerOutput-setSizePolicy() with each of its options...nothing so far has affected the size. Like I said, I would prefer a simple solution in a few lines of code to having to create a sub-class just to change sizeHint(). All suggestions are appreciated.

    Read the article

  • JEE Web Applications vs Web Services

    - by Zac
    Can someone confirm or clarify for me: From what I can tell, JEE web apps consist of a Servlet and/or JSP driven dynamic web page being fed back in the HTTP response, triggered by the JEE server receiving a HTTP GET or POST request. From what I can tell, JEE web services also make use of Servlets as the web tier components, however a WS Servlet receives a SOAP message and validates the contents of those messages with whatever WSDL the Servlet is WARed with. The response is also packaged in SOAP and sent back to the requestor. So, from what I can tell, both JEE web apps and WSes use Servlets as the web components, with the only real difference being the protocol used (raw HTTP vs SOAP, which is an extension of HTTP). This is the best I could come up with - am I right? Totally wrong? Close?

    Read the article

  • Help with a pseudo-class, first-child

    - by zac
    I am working on a CMS platform with limited access to the template files and want to try and control some of the layout with pseudo class but no luck yet. Can anyone see what is wrong with this structure and why my pseudo class is being ignored? <div id="main"> <div class="someRandomDiv"></div> <div class="block"> stuff </div> <div class="block"> more stuff </div> </div> and i am trying something like this #main .block {border: 1px solid blue} #main .block:first-child {border: 1px solid red} so with this example I would think the stuff block would have a red border and more stuff would have a blue but it is all just blue. Thanks for any help with this.

    Read the article

  • Preserve onchange for a dropdown list when setting the value with Javascript.

    - by Zac Altman
    I have a dropdown list with a piece of code that is run when the value is changed: <select name="SList" onchange="javascript:setTimeout('__doPostBack(\'SList\',\'\')', 0)" id="SList"> Everything works fine when manually done. As an option is selected, the onchange code is called. The problem begins when I try to change the selected value using a piece of Javscript. I want to be able to automatically change the selected option using JS, whilst still having the onchange code called, exactly as if done manually. I try calling this: form.SList.value = "33"; The right option is selected, but the onchange code does not get called. So then I try doing this: form.SList.value = "33"; javascript:setTimeout('__doPostBack(\'SList\',\'\')', 0); The right value is not selected and nothing happens. FYI, the code is done in ASP.NET and Javascript. What can I run to change the selected option whilst still calling the onchange code?

    Read the article

< Previous Page | 1 2 3 4 5 6 7 8 9 10  | Next Page >