Search Results

Search found 32994 results on 1320 pages for 'second level cache'.

Page 87/1320 | < Previous Page | 83 84 85 86 87 88 89 90 91 92 93 94  | Next Page >

  • Looking for a better design: A readonly in-memory cache mechanism

    - by Dylan Lin
    Hi all, I have a Category entity (class), which has zero or one parent Category and many child Categories -- it's a tree structure. The Category data is stored in a RDBMS, so for better performance, I want to load all categories and cache them in memory while launching the applicaiton. Our system can have plugins, and we allow the plugin authors to access the Category Tree, but they should not modify the cached items and the tree(I think a non-readonly design might cause some subtle bugs in this senario), only the system knows when and how to refresh the tree. Here are some demo codes: public interface ITreeNode<T> where T : ITreeNode<T> { // No setter T Parent { get; } IEnumerable<T> ChildNodes { get; } } // This class is generated by O/R Mapping tool (e.g. Entity Framework) public class Category : EntityObject { public string Name { get; set; } } // Because Category is not stateless, so I create a cleaner view class for Category. // And this class is the Node Type of the Category Tree public class CategoryView : ITreeNode<CategoryView> { public string Name { get; private set; } #region ITreeNode Memebers public CategoryView Parent { get; private set; } private List<CategoryView> _childNodes; public IEnumerable<CategoryView> ChildNodes { return _childNodes; } #endregion public static CategoryView CreateFrom(Category category) { // here I can set the CategoryView.Name property } } So far so good. However, I want to make ITreeNode interface reuseable, and for some other types, the tree should not be readonly. We are not able to do this with the above readonly ITreeNode, so I want the ITreeNode to be like this: public interface ITreeNode<T> { // has setter T Parent { get; set; } // use ICollection<T> instead of IEnumerable<T> ICollection<T> ChildNodes { get; } } But if we make the ITreeNode writable, then we cannot make the Category Tree readonly, it's not good. So I think if we can do like this: public interface ITreeNode<T> { T Parent { get; } IEnumerable<T> ChildNodes { get; } } public interface IWritableTreeNode<T> : ITreeNode<T> { new T Parent { get; set; } new ICollection<T> ChildNodes { get; } } Is this good or bad? Are there some better designs? Thanks a lot! :)

    Read the article

  • How to return a property name when comparing two properties at class-level

    - by CodeMonkey
    Hi I have implemented an 'EqualTo' Validation Attribute, that compares two Properties of an object, during ModelBinding in ASP.NET MVC 2. The problem I have is not with it not working, because it does work. The problem is, when I do my request - which is an ajax request - I get back errors to my front-end, where it sets a class on the input fields to indicate invalid input. What it does is iterate through a list of Errors (in a JsonResult), and set a class. This is all dandy. But the ValidationAtrribute I am having trouble with is set at a Class-level, i.e., it's not like other ValidationAttributes where you set something like "[Required]" or something like that. [AttributeUsage(AttributeTargets.Class, AllowMultiple=true, Inherited=false)] public class EqualToAttribute : ValidationAttribute { public String SourceProperty { get; set; } public String MatchProperty { get; set; } public EqualToAttribute(string source, string match) { SourceProperty = source; MatchProperty = match; } public override Boolean IsValid(Object value) { Type objectType = value.GetType(); PropertyInfo[] properties = objectType.GetProperties(); object sourceValue = new object(); object matchValue = new object(); Type sourceType = null; Type matchType = null; int counter = 0; foreach (PropertyInfo propertyInfo in properties) { if (propertyInfo.Name == SourceProperty || propertyInfo.Name == MatchProperty) { if (counter == 0) { sourceValue = propertyInfo.GetValue(value, null); sourceType = propertyInfo.GetValue(value, null).GetType(); } if (counter == 1) { matchValue = propertyInfo.GetValue(value, null); matchType = propertyInfo.GetValue(value, null).GetType(); } counter++; if (counter == 2) { break; } } } if (sourceType != null && matchType != null) { return sourceValue.ToString().Equals(matchValue.ToString()); //return Convert.ChangeType(sourceValue, sourceType) == Convert.ChangeType(matchValue, matchType); } return false; } private object _typeId = new object(); public override object TypeId { get { return this._typeId; } } } Now this code works, except for the fact that the validation process does not return which Property failed. And I simply can't figure out how to make it return one of the two. In reality I don't care which one it returns.. because both are failing.. Do you have an idea how to make it return the/or both Property/Properties that is/are failing.

    Read the article

  • Getting level values from PCM raw data using Core Audio

    - by John
    I am trying to extract level data from a PCM audio file using core audio. I have gotten as far as (I believe) getting the raw data into a byte array (UInt8) but it is 16 bit PCM data and I am having trouble reading the data out. The input is from the iPhone microphone, which I have set as: [recordSetting setValue:[NSNumber numberWithInt:kAudioFormatLinearPCM] forKey:AVFormatIDKey]; [recordSetting setValue:[NSNumber numberWithFloat:44100.0] forKey:AVSampleRateKey]; [recordSetting setValue:[NSNumber numberWithInt:1] forKey:AVNumberOfChannelsKey]; [recordSetting setValue:[NSNumber numberWithInt:16] forKey:AVLinearPCMBitDepthKey]; [recordSetting setValue:[NSNumber numberWithBool:NO] forKey:AVLinearPCMIsBigEndianKey]; [recordSetting setValue:[NSNumber numberWithBool:NO] forKey:AVLinearPCMIsFloatKey]; which is obviously 16 bits. I am then trying to just print out a few values to see if they look reasonable for debug purposes below, and they do not look reasonable (many 0's). ExtAudioFileRef inputFile = NULL; ExtAudioFileOpenURL(track.location, &inputFile); AudioStreamBasicDescription inputFileFormat; UInt32 dataSize = (UInt32)sizeof(inputFileFormat); ExtAudioFileGetProperty(inputFile, kExtAudioFileProperty_FileDataFormat, &dataSize, &inputFileFormat); UInt8 *buffer = malloc(BUFFER_SIZE); AudioBufferList bufferList; bufferList.mNumberBuffers = 1; bufferList.mBuffers[0].mNumberChannels = 1; bufferList.mBuffers[0].mData = buffer; //pointer to buffer of audio data bufferList.mBuffers[0].mDataByteSize = BUFFER_SIZE; //number of bytes in the buffer while(true) { UInt32 frameCount = (bufferList.mBuffers[0].mDataByteSize / inputFileFormat.mBytesPerFrame); // Read a chunk of input OSStatus status = ExtAudioFileRead(inputFile, &frameCount, &bufferList); // If no frames were returned, conversion is finished if(0 == frameCount) break; NSLog(@"---"); int16_t *bufferl = &buffer; for(int i=0;i<100;i++){ //const int16_t *bufferl = bufferl[i]; NSLog(@"%d",bufferl[i]); } } Not sure what I am doing wrong, I think it has to do with reading the byte array. Sorry for the long code post...

    Read the article

  • Cannot use READPAST in snapshot isolation mode

    - by Marcus
    I have a process which is called from multiple threads which does the following: Start transaction Select unit of work from work table with by finding the next row where IsProcessed=0 with hints (UPDLOCK, HOLDLOCK, READPAST) Process the unit of work (C# and SQL stored procedures) Commit the transaction The idea of this is that a thread dips into the pool for the "next" piece of work, and processes it, and the locks are there to ensure that a single piece of work is not processed twice. (the order doesn't matter). All this has been working fine for months. Until today that is, when I happened to realise that despite enabling snapshot isolation and making it the default at the database level, the actual transaction creation code was manually setting an isolation level of "ReadCommitted". I duly changed that to "Snapshot", and of course immediately received the "You can only specify the READPAST lock in the READ COMMITTED or REPEATABLE READ" error message. Oops! The main reason for locking the row was to "mark the row" in such a way that the "mark" would be removed when the transaction that applied the mark was committed and the lock seemed to be the best way to do this, since this table isn't read otherwise except by these threads. If I were to use the IsProcessed flag as the lock, then presumably I would need to do the update first, and then select the row I just updated, but I would need to employ the NOLOCK flag to know whether any other thread had set the flag on a row. All sounds a bit messy. The easiest option would be to abandon the snapshot isolation mode altogether, but the design of step #3 requires it. Any bright ideas on the best way to resolve this problem? Thanks Marcus

    Read the article

  • having trouble with jpa, looks to be reading from cache when told to ignore

    - by jeff
    i'm using jpa and eclipselink. it says version 2.0.2.v20100323-r6872 for eclipselink. it's an SE application, small. local persistence unit. i am using a postgres database. i have some code the wakes up once per second and does a jpa query on a table. it's polling to see whether some fields on a given record change. but when i update a field in sql, it keeps showing the same value each time i query it. and this is after waiting, creating a new entitymanager, and giving a query hint. when i query the table from sql or python, i can see the changed field. but from within my jpa query, once i've read the value, it never changes in later executions of the query -- even though the entity manager has been recreated. i've tried telling it to not look in the cache. but it seems to ignore that. very puzzling, can you help please? here is the method. i call this once every 3 seconds. the tgt.getTrlDlrs() println shows me i get the same value for this field on every call of the method ("value won't change"). even if i change the field in the database. when i query the record from outside java, i see the change right away. also, if i stop the java program and restart it, i see the new value printed out immediately: public void exitChk(EntityManagerFactory emf, Main_1 mn){ // this will be a list of the trade close targets that are active List<Pairs01Tradeclosetgt> mn_res = new ArrayList<Pairs01Tradeclosetgt>(); //this is a list of the place number in the array list above //of positions to close ArrayList<Integer> idxsToCl = new ArrayList<Integer>(); EntityManager em = emf.createEntityManager(); em.getTransaction().begin(); em.clear(); String qryTxt = "SELECT p FROM Pairs01Tradeclosetgt p WHERE p.isClosed = :isClosed AND p.isFilled = :isFilled"; Query qMn = em.createQuery(qryTxt); qMn.setHint(QueryHints.CACHE_USAGE, CacheUsage.DoNotCheckCache); qMn.setParameter("isClosed", false); qMn.setParameter("isFilled", true); mn_res = (List<Pairs01Tradeclosetgt>) qMn.getResultList(); // are there any trade close targets we need to check for closing? if (mn_res!=null){ //if yes, see whether they've hit their target for (int i=0;i<mn_res.size();i++){ Pairs01Tradeclosetgt tgt = new Pairs01Tradeclosetgt(); tgt = mn_res.get(i); System.out.println("value won't change:" + tgt.getTrlDlrs()); here is my entity class (partial): @Entity @Table(name = "pairs01_tradeclosetgt", catalog = "blah", schema = "public") @NamedQueries({ @NamedQuery(name = "Pairs01Tradeclosetgt.findAll", query = "SELECT p FROM Pairs01Tradeclosetgt p"), @NamedQuery(name = "Pairs01Tradeclosetgt.findByClseRatio", query = "SELECT p FROM Pairs01Tradeclosetgt p WHERE p.clseRatio = :clseRatio")}) public class Pairs01Tradeclosetgt implements Serializable { private static final long serialVersionUID = 1L; @Id @Basic(optional = false) @Column(name = "id") @SequenceGenerator(name="pairs01_tradeclosetgt_id_seq", allocationSize=1) @GeneratedValue(strategy=GenerationType.SEQUENCE, generator = "pairs01_tradeclosetgt_id_seq") private Integer id; and my persitence unit: <?xml version="1.0" encoding="UTF-8"?> <persistence version="1.0" xmlns="http://java.sun.com/xml/ns/persistence" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://java.sun.com/xml/ns/persistence http://java.sun.com/xml/ns/persistence/persistence_1_0.xsd"> <persistence-unit name="testpu" transaction-type="RESOURCE_LOCAL"> <provider>org.eclipse.persistence.jpa.PersistenceProvider</provider> <class>mourv02.Pairs01Quotes</class> <class>mourv02.Pairs01Pair</class> <class>mourv02.Pairs01Trderrs</class> <class>mourv02.Pairs01Tradereq</class> <class>mourv02.Pairs01Tradeclosetgt</class> <class>mourv02.Pairs01Orderstatus</class> <properties> <property name="javax.persistence.jdbc.url" value="jdbc:postgresql://192.168.1.80:5432/blah"/> <property name="javax.persistence.jdbc.password" value="secret"/> <property name="javax.persistence.jdbc.driver" value="org.postgresql.Driver"/> <property name="javax.persistence.jdbc.user" value="noone"/> </properties> </persistence-unit> </persistence>

    Read the article

  • IIS7 Network Bios command limit and disabling file change notifications.

    - by meandmycode
    Having experienced IIS for many years I've hit this error various times in the past, and understand it, the gist being IIS is attempting to setup so many file change notifications over a UNC share that it is hitting a specified limit of active commands. However, whilst I understand this is a registry change on the endpoints, in our internal network we're starting to hit this when running certain domains that have large content (and the content is then stored out of source control and instead on a dumb network store). Given this is a testing environment and rather than changing registry settings (which may not even work for client (dev) machines that run Vista IIS), I'd like to stop IIS setting up all these file change notifications. Is this possible to do on a per site basis? additionally the configuration can't be in the web.config, if this means having to disable FCN's for the entire IIS server this would be preferred. Thanks in advance.

    Read the article

  • If-Modified-Since vs If-None-Match

    - by Roger
    This question is based on this article response header HTTP/1.1 200 OK Last-Modified: Tue, 12 Dec 2006 03:03:59 GMT ETag: "10c24bc-4ab-457e1c1f" Content-Length: 12195 request header GET /i/yahoo.gif HTTP/1.1 Host: us.yimg.com If-Modified-Since: Tue, 12 Dec 2006 03:03:59 GMT If-None-Match: "10c24bc-4ab-457e1c1f" HTTP/1.1 304 Not Modified In this case browser is sending both If-None-Match and If-Modified-Since. My question is on the server side do I need to match BOTH etag and If-Modified-Since before I send 304. Or Should I just look at etag and send 304 if etag is a match. In this case I am ignoring If-Modified-Since .

    Read the article

  • MemCache-repcached compile error

    - by Ramy Allam
    I'm trying to install [memcached-1.2.8-repcached-2.2.1]( http://sourceforge.net/projects/repcached/files/latest/download?source=files) And I have the following error after running the make command: make all-recursive make[1]: Entering directory `/usr/local/src/pro/memcached-1.2.8-repcached-2.2.1' Making all in doc make[2]: Entering directory `/usr/local/src/pro/memcached-1.2.8-repcached-2.2.1/doc' make[2]: Nothing to be done for `all'. make[2]: Leaving directory `/usr/local/src/pro/memcached-1.2.8-repcached-2.2.1/doc' make[2]: Entering directory `/usr/local/src/pro/memcached-1.2.8-repcached-2.2.1' gcc -DHAVE_CONFIG_H -I. -DNDEBUG -g -O2 -MT memcached-memcached.o -MD -MP -MF .d eps/memcached-memcached.Tpo -c -o memcached-memcached.o test -f 'memcached.c' || echo './'memcached.c memcached.c: In function ‘add_iov’: memcached.c:697: error: ‘IOV_MAX’ undeclared (first use in this function) memcached.c:697: error: (Each undeclared identifier is reported only once memcached.c:697: error: for each function it appears in.) make[2]: * [memcached-memcached.o] Error 1 make[2]: Leaving directory `/usr/local/src/pro/memcached-1.2.8-repcached-2.2.1' make[1]: * [all-recursive] Error 1 make[1]: Leaving directory `/usr/local/src/pro/memcached-1.2.8-repcached-2.2.1' make: * [all] Error 2 OS : Centos5.7 64bit gcc-4.1.2-51.el5 gcc-c++-4.1.2-51.el5 libgcc-4.1.2-51.el5 Note : Memcached and memcache extension for php are already installed root@server[~]# memcached -h memcached 1.4.5 php ext http://pecl.php.net/get/memcache-2.2.6.tgz

    Read the article

  • Stop Windows Domain Environment Caching Old Passwords?

    - by Daryl Gill
    I have noticed on my domain environment; the old Administrator password (before password expire).. The client machines have cached the old password and have the ability to bypass the new password by entering the old one..? I have noticed on my domain environment; the old Administrator password (before password expire).. The client machines have cached the old password and have the ability to bypass the new password by entering the old one..? Basically; I'm running a UAC enabled domain, which needs the administrator password to continue basic stuff; installations and such. The password for the administrator account has been changed due to expiration of said accounts password. By accident a fellow administrator typed the old password and still bypassed the UAC with what should have been the incorrect password. Is this a bug with the environment? or something that needs to be tweaked in the server sided settings? Is this a bug with the environment? or something that needs to be tweaked in the server sided settings?

    Read the article

  • Squid 3 reloading makes it stop serving requests

    - by coredump
    So, we use Squid 3 here (3.0.STABLE8-3+lenny4), pretty standard configuration (no dansguardian or similar) + NTLM authentication with LDAP background, circa 1000 users on a busy day, and our acls reference some external files (allowed/blocked sites/ip addresses). On Squid 2.X we used to be capable of reloading it's configuration (to add or sites or addresses to rules, etc) and squid would not stop serving during the reload. Since we changed to 3.0, that seems to be impossible: everytime we use reload (or -k reconfigure) it stop serving requests for as long as 2 minutes, and clients receive a Configured proxy is not accepting connections message. I checked the documentation and got nothing about it, does anyone else suffer from this problem or is it a isolated case on my setup? Also, if you have Squid 3.0 and doesn't suffer from this problem, how is your squid configured?

    Read the article

  • Disable mod_cache to fix website

    - by Zach Russell
    I am working on a website that runs on WordPress. The an error occurs when accessing the blog page that it displays the RSS feed (in XML) rather than the posts. After some research I found out the when enabling mod_cache this error sometimes occurs. I see when looking at the apache2 modules enabled apache2ctl -M it shows that mod_cache, mod_mem_cache and mod_disc_cache are enabled. What should I disable in order to resolve this problem?

    Read the article

  • Cisco WebVPN RDP Plugin and NLA

    - by bab
    I'm having trouble finding anything in Cisco's docs or with Google searches, so I'm hoping someone out in ServerFault land might know. We've recently enabled NLA domain-wide to protect against some of the recent RDP vulnerabilities. However, we can no longer use the Cisco WebVPN on our ASA to connect to these boxes (Connection Failure). I assume this is because the RDP2 plugin (as of Apr 27 2012) doesn't support NLA? Is there another version of the plugin that does? Thanks!

    Read the article

  • Slow Memcached: Average 10ms memcached `get`

    - by Chris W.
    We're using Newrelic to measure our Python/Django application performance. Newrelic is reporting that across our system "Memcached" is taking an average of 12ms to respond to commands. Drilling down into the top dozen or so web views (by # of requests) I can see that some Memcache get take up to 30ms; I can't find a single use of Memcache get that returns in less than 10ms. More details on the system architecture: Currently we have four application servers each of which has a memcached member. All four memcached members participate in a memcache cluster. We're running on a cloud hosting provider and all traffic is running across the "internal" network (via "internal" IPs) When I ping from one application server to another the responses are in ~0.5ms Isn't 10ms a slow response time for Memcached? As far as I understand if you think "Memcache is too slow" then "you're doing it wrong". So am I doing it wrong? Here's the output of the memcache-top command: memcache-top v0.7 (default port: 11211, color: on, refresh: 3 seconds) INSTANCE USAGE HIT % CONN TIME EVICT/s GETS/s SETS/s READ/s WRITE/s cache1:11211 37.1% 62.7% 10 5.3ms 0.0 73 9 3958 84.6K cache2:11211 42.4% 60.8% 11 4.4ms 0.0 46 12 3848 62.2K cache3:11211 37.5% 66.5% 12 4.2ms 0.0 75 17 6056 170.4K AVERAGE: 39.0% 63.3% 11 4.6ms 0.0 64 13 4620 105.7K TOTAL: 0.1GB/ 0.4GB 33 13.9ms 0.0 193 38 13.5K 317.2K (ctrl-c to quit.) ** Here is the output of the top command on one machine: ** (Roughly the same on all cluster machines. As you can see there is very low CPU utilization, because these machines only run memcache.) top - 21:48:56 up 1 day, 4:56, 1 user, load average: 0.01, 0.06, 0.05 Tasks: 70 total, 1 running, 69 sleeping, 0 stopped, 0 zombie Cpu(s): 0.0%us, 0.0%sy, 0.0%ni, 99.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.3%st Mem: 501392k total, 424940k used, 76452k free, 66416k buffers Swap: 499996k total, 13064k used, 486932k free, 181168k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 6519 nobody 20 0 384m 74m 880 S 1.0 15.3 18:22.97 memcached 3 root 20 0 0 0 0 S 0.3 0.0 0:38.03 ksoftirqd/0 1 root 20 0 24332 1552 776 S 0.0 0.3 0:00.56 init 2 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kthreadd 4 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kworker/0:0 5 root 20 0 0 0 0 S 0.0 0.0 0:00.02 kworker/u:0 6 root RT 0 0 0 0 S 0.0 0.0 0:00.00 migration/0 7 root RT 0 0 0 0 S 0.0 0.0 0:00.62 watchdog/0 8 root 0 -20 0 0 0 S 0.0 0.0 0:00.00 cpuset 9 root 0 -20 0 0 0 S 0.0 0.0 0:00.00 khelper ...output truncated...

    Read the article

  • Caching API Proxy Server

    - by edc1591
    I need to have a server that caches API responses and then forwards them along to a desktop app. I don't really have much experience with this, so I have a few questions. First of all, what kind of server should I get? I already use Linode for my websites, so ideally I'd like to go with them. I expect to get anywhere from 30 million to 40 million requests to my proxy server each month. Will a 512 Linode be able to support that? Also, is there any software out there that does this already, or will I have to write my own? The API responses are roughly 10 KB each on average, so doing the math, that's a lot of data each month. Should I just add more transfer to whatever server I buy, or can I somehow compress the API responses before sending them off to the user? Thanks for any help.

    Read the article

  • Ignoring GET parameters in Varnish VCL

    - by JamesHarrison
    Okay: I've got a site set up which has some APIs we expose to developers, which are in the format /api/item.xml?type_ids=34,35,37&region_ids=1000002,1000003&key=SOMERANDOMALPHANUM In this URI, type_ids is always set, region_ids and key are optional. The important thing to note is that the key variable does not affect the content of the response. It is used for internal tracking of requests so we can identify people who make slow or otherwise unwanted requests. In Varnish, we have a VCL like this: if (req.http.host ~ "the-site-in-question.com") { if (req.url ~ "^/api/.+\.xml") { unset req.http.cookie; } } We just strip cookies out and let the backend do the rest as far as times are concerned (this is a hackaround since Rails/authlogic sends session cookies with API responses). At present though, any distinct developers are basically hitting different caches since &key=SOMEALPHANUM is considered as part of the Varnish hash for storage. This is obviously not a great solution and I'm trying to work out how to tell Varnish to ignore that part of the URI.

    Read the article

  • SQL Service Broker enabled causes 100% CPU

    - by user40373
    I have new set of code for a website that is using SqlCacheDependencies based on sql commands. I have enabled SQL Service Broker and some triggers on update/insert/delete and it is causing 100% CPU. Any ideas if I am doing something wrong or suggestions to improve? Here are the SQLchanges I ran: alter database DATABASE_NAME set enable_broker WITH ROLLBACK IMMEDIATE grant subscribe query notifications to CONNECTION_USER_NAME grant send on service::sqlquerynotificationservice to CONNECTION_USER_NAME ALTER AUTHORIZATION ON DATABASE::DATABASE_NAME TO CONNECTION_USER_NAME;

    Read the article

  • Linux not buffering block I/O when the device is not "in use" (i.e. mounted)

    - by Radek Hladík
    I am installing new server and I've found an interesting issue. The server is running Fedora 19 (3.11.7-200.fc19.x86_64 kernel) and is supposed to host a few KVM/Qemu virtual servers (mail server, file server, etc..). The HW is Intel(R) Xeon(R) CPU 5160 @ 3.00GHz with 16GB RAM. One of the most important features will be Samba server and we have decided to make it as virtual machine with almost direct access to the disks. So the real HDD is cached on SSD (via bcache) then raided with md and the final device is exported into the virtual machine via virtio. The virtual machine is again Fedora 19 with the same kernel. One important topic to find out is whether the virtualization layer will not introduce high overload into disk I/Os. So far I've been able to get up to 180MB/s in VM and up to 220MB/s on real HW (on the SSD disk). I am still not sure why the overhead is so big but it is more than the network can handle so I do not care so much. The interesting thing is that I've found that the disk reads are not buffered in the VM unless I create and mount FS on the disk or I use the disks somehow. Simply put: Lets do dd to read disk for the first time (the /dev/vdd is an old Raptor disk 70MB/s is its real speed): [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 36.8038 s, 71.2 MB/s Buffers: 14444 kB Rereading the data shows that they are cached somewhere but not in buffers of the VM. Also the speed increased to "only" 500MB/s. The VM has 4GB of RAM (more that the test file) [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 5.16016 s, 508 MB/s Buffers: 14444 kB [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 5.05727 s, 518 MB/s Buffers: 14444 kB Now lets mount the FS on /dev/vdd and try the dd again: [root@localhost ~]# mount /dev/vdd /mnt/tmp [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 4.68578 s, 559 MB/s Buffers: 2574592 kB [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 1.50504 s, 1.7 GB/s Buffers: 2574592 kB While the first read was the same, all 2.6GB got buffered and the next read was at 1.7GB/s. And when I unmount the device: [root@localhost ~]# umount /mnt/tmp [root@localhost ~]# cat /proc/meminfo | grep Buffers Buffers: 14452 kB [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 5.10499 s, 514 MB/s Buffers: 14468 kB The bcache was disabled while testing and the results are same on faster (newer) HDDs and on SSD (except for the initial read speed of course). To sum it up. When I read from the device via dd first time, it gets read from the disk. Next time I reread it gets cached in the host but not in the guest (thats actually the same issue, more on that later). When I mount the filesystem but try to read the device directly it gets cached in VM (via buffers). As soon as I stop "using" it, buffers are discarded and the device is not cached anymore in the VM. When I looked into buffers value on the host I realized that the situation is the same. The block I/O gets buffered only when the disk is in use, in this case it means "exported to a VM". On host, after all the measurement done: 3165552 buffers On the host, after the VM shutdown: 119176 buffers I know it is not important as the disks will be mounted all the time but I am curious and I would like to know why it is working like this.

    Read the article

  • Per-machine decentralised DNS caching - nscd/lwresd/etc

    - by Dan Carley
    Preface: We have caching resolvers at each of our geographic network locations. These are clustered for resiliency and their locality reduces the latency of internal requests generated by our servers. This works well. Except that a vast quantity of the requests seen over the wire are lookups for the same records, generated by applications which don't perform any DNS caching of their own. Questions: Is there a significant benefit to running lightweight caching daemons on the individual servers in order to reduce repeated requests from hitting the network? Does anyone have experience of using [u]nscd, lwresd or dnscache to do such a thing? Are there any other packages worth looking at? Any caveats to beware of? Besides the obvious, caching and negative caching stale results.

    Read the article

  • postfix concurrency limit with round robin dns

    - by goose
    Take the following internal round robin dns setup mymta.com. IN A 172.31.1.1 mymta.com. IN A 172.31.1.2 mymta.com. IN A 172.31.1.3 mymta.com. IN A 172.31.1.4 mymta.com. IN A 172.31.1.5 mymta.com. IN A 172.31.1.6 mymta.com. IN A 172.31.1.7 mymta.com. IN A 172.31.1.8 mymta.com. IN A 172.31.1.9 mymta.com. IN A 172.31.1.10 Now assume the following postfix setup (assume these are the only tweaks from defaults in debian package) main.cf: smtp_connection_cache_destinations = mymta.com smtp_connection_cache_reuse_limit = 750 smtp_destination_concurrency_limit = 75 transport * :[mymta.com] I would expect 75 concurrent connections spread across the 10 A records I've set in DNS. However I'm seeing more than a few hundred connections to mymta.com and I'm wondering if Postfix is "smart" enough to set up 75 concurrent connections for each IP address. Thoughts?

    Read the article

  • IIS seems to be caching files on a system share?

    - by scott novell
    Switching over to windows 2008 and IIS 7.5 and it seems whenever I make a change to a css file on a system share it does not show through the browser for a few mins. It is shown through the browser using an ISAPI filter. I have turned off output caching in IIS and also turned off caching on the share itself. The browser is not caching either forcing a 200 and it is cached. Any ideas

    Read the article

  • If-Modified-Since vs If-None-Match

    - by Roger
    This question is based on this article response header HTTP/1.1 200 OK Last-Modified: Tue, 12 Dec 2006 03:03:59 GMT ETag: "10c24bc-4ab-457e1c1f" Content-Length: 12195 request header GET /i/yahoo.gif HTTP/1.1 Host: us.yimg.com If-Modified-Since: Tue, 12 Dec 2006 03:03:59 GMT If-None-Match: "10c24bc-4ab-457e1c1f" HTTP/1.1 304 Not Modified In this case browser is sending both If-None-Match and If-Modified-Since. My question is on the server side do I need to match BOTH etag and If-Modified-Since before I send 304. Or Should I just look at etag and send 304 if etag is a match. In this case I am ignoring If-Modified-Since .

    Read the article

  • Squid closing the connection on long HTTP GET requests

    - by Rhys
    Hello, When running a database query on a specific external site we use, Squid seems to cut off the connection after a consistent period of time (just over a minute). The query is submitted through a standard web form is that uses GET to query their database. Firefox 3 just displays a blank page. Internet Explorer throws a 'Page Cannot Be Displayed' error (tested in v6 and v8). When we perform the same query on the same machine, but bypass the Squid proxy, it works fine. The query takes about two and a half minutes to complete. There are a few timeout settings in Squid, but I honestly don't know what one to be looking at. Any possible solutions would be much appreciated. Cheers

    Read the article

< Previous Page | 83 84 85 86 87 88 89 90 91 92 93 94  | Next Page >