Search Results

Search found 1210 results on 49 pages for 'notifications'.

Page 28/49 | < Previous Page | 24 25 26 27 28 29 30 31 32 33 34 35  | Next Page >

• SMTP server problem

  - by ram

  Hi, Our requirement is to send weekly newsletters to our website customers. For which we wanted to have local hosted SMTP server in our office. We are not using SMTP server provided by website hosting provider, as we wanted to reduce the network traffic and avoid IP blocking due to bulk mails. We are sending newsletters on weekly basis from our local SMTP server. But due to some reasons, some emails are going to spam and some are not reaching to customers and sometimes there are bounce messages to follow bulk email guidelines (mainly from Gmail). Can you please suggest me, how to achieve my problem. I also wanted to know what type of technology generally Linkedin or banks uses to send notifications emails to all its customers. When they send bulk emails, they will always reach inbox with out any problem. I want the same solution to implement for my website. Please suggest me. Thank you very much in advance.

  Read the article

• Change popup format for Kopete in Gnome (Ubuntu 10.04)

  - by HorusKol

  After trialling Empathy which was included with Gnome/Ubuntu 10.04 I decided to go back to Kopete since I was losing chat messages. However, for some reason, Kopete is now displaying all popups in a big, ugly window with four options (ok, cancel, view, ignore, or something like that) that actually all seem to do nothing. I tried looking for options in Kopete to change the popup style to the nicer Gnome style one which goes up in the top-right of my desktop, but it doesn't seem to be possible with this release. So I had to resort to removing all popup messages from Kopete to prevent my screen being taking over by a popup requiring action for every incoming chat message. Unfortunately, this now means that I do not get any notifications when people message me - and so I can go a couple of hours without realising that they did. Anyone know how to get the nice, unobtrusive popups working in this version?

  Read the article

• Is there a way to have customised text instead of subject in Thunderbird's inbox list?

  - by peterp

  I am getting a lot of informational emails like "You've got a new message from ..." or "Notification of Donation Received", which often do not contain any information in the subject so that I have to open the email to see who sent the message or who donated which amount. I'd love to be able to make TB parse incoming emails and then display something interesting instead of the original subject, e.g. by defining a regular expression pattern. I know how to write regular expressions, but I do not know whether there is a way or an addon to modify the displayed text in the messages view. EDIT for clarification: I would like donation notifications from Paypal not to be displayed as original Notification of Donation Received but rather Paypal: John Doe has donated 50$

  Read the article

• diagnose "corrupt file" problems

  - by Matthew

  My computer has been crashing the last couple weeks pretty regularly (at least once a day). A lot of times things I do will display a little notification in the bottom right saying something about a corrupt file. (I'm on Windows XP Pro Service Pack 3). When the computer does crash I get the "blue screen of death" usually. Some of the notifications also advise running the chkdsk utility. I cannot get it to successfully run. Using the command prompt (or even the "tools" menu after right clicking the drive and choosing properties), it will not run the utility (it says "do you want to schedule it to run next boot time" or whatever, which I confirm). The problem is that most of the time after restarting, it doesn't run at all. The few times it does run, it has an error (I can't remember the error right now, it at least says it's ntfs and such) and says disk checking will end. How can I get it to successfully run?

  Read the article

• Windows Updates settings issue!

  - by Kumi

  Hi, it's my first post here, don't wanna make it long, but I got a problem with my 'Windows Update' it won't show its setting's options. It only shows these settings: Important Updates, Recommended Updates and Who can install updates, it removed the other setting sections like 'Microsoft Updates' "Give me updates for Microsoft products and check for new optional Microsoft software when I update windows" or the last settings section 'Software notification' "Show me detailed notifications when new Microsoft software is available" Here's a screenshot of my Windows Update: http://img96.imageshack.us/img96/1963/windowsupdatescreenshot.png And here's how it was before and how should be: h**p://www.sevenforums.com/attachments/tutorials/3523d1242168648-windows-update-settings-change-wu_settings.jpg Thanks. PS: I'm running windows 7 RTM

  Read the article

• Desktop notifcations for IMAP subfolders in Outlook 2013

  - by hplieninger

  I use Outlook 2013 and an IMAP account for my mails. I have configured several filters (rules) directly in the webmail application of my provider in order to deliver certain mails, e.g., from my boss, into certain folders other then the "inbox". Note that I did not define any rules in Outlook itself and I also do not want to do so (because I want these mails being delivered into the folders not only in Outlook but also on other devices). The problem is that that I receive desktop notifications only for incoming mails in the main folder ("Inbox") but not for mails in any of the other folders.

  Read the article

• Change popup format for Kopete in Gnome (Ubuntu 10.04)

  - by HorusKol

  After trialling Empathy which was included with Gnome/Ubuntu 10.04 I decided to go back to Kopete since I was losing chat messages. However, for some reason, Kopete is now displaying all popups in a big, ugly window with four options (ok, cancel, view, ignore, or something like that) that actually all seem to do nothing. I tried looking for options in Kopete to change the popup style to the nicer Gnome style one which goes up in the top-right of my desktop, but it doesn't seem to be possible with this release. So I had to resort to removing all popup messages from Kopete to prevent my screen being taking over by a popup requiring action for every incoming chat message. Unfortunately, this now means that I do not get any notifications when people message me - and so I can go a couple of hours without realising that they did. Anyone know how to get the nice, unobtrusive popups working in this version?

  Read the article

• Force a Windows 7 user to change password when it has expired

  - by Joshua King

  Is it possible to force users (on Windows 7 PCs) to change their password once it has expired? Currently our users get notifications that their passwords will expire in 7, 3, 2 and 1 days but once the password has expired they are still able to log into their machines with the expired password. Most of our users ignore the warnings and don't change their password until we tell them they have to (when they ring Help Desk asking why their email and intranet isn't working). From memory, Windows XP machines would show a message at login that the password had expired and then offer the change password screen. Is it possible to do the same thing under Win7? Similar to how if you set a user's account to "must change password at next login" it gives the change password screen automatically.

  Read the article

• Windows webserver monitor and notifier service

  - by WestDiscGolf

  I'm looking for a software tool/service, preferably opensource/free, which will run on a Windows 2003 r3 standard server which will monitor bandwidth usage of all the websites it's hosting (a break down by site would be very useful) and also send out notifications if IIS or another service (as defined) stops responding. A web interface to see the bandwidth usage and other details would be very useful. I've personally got a vps which does this kinda of monitoring but I believe this is more linked into the host itself. Does anyone have any pointers? All help is appriciated. Cheers

  Read the article

• How should I troubleshoot a problematic wireless connection on Linux?

  - by Gearoid Murphy

  I recently purchased a netgear 150 usb wireless dongle for use with my 11.10 Xubuntu amd64 system. Using the network-manager interface, I can see local wireless networks and enter the authentication details for my local wireless lan. Unfortunately, the connection does not seem to work, I keep getting notifications that my wireless has disconnected (but none indicating that I've connected). When I examine syslog, it seems to indicate that I've successfully associated with the wireless switch and that dhcp has successfully acquired an ip address but the log shows that the dhcp process keeps sending requests, eventually dropping the connection. 'ifconfig wlan0' never shows the dhcp address logged in syslog. I suspect that the problem lies with the usb dongle, my configuration or the wireless switch but I am not certain how to isolate the problem, can anyone provide some insight on how I should go about homing in on the cause of this problem or verifying the functionality of the individual components, thanks.

  Read the article

• Multi-account Google Apps Calendar Free-busy status?

  - by Andrew Bolster

  I have 3 google apps domain accounts and a personal google account, and until recently, there was little need for the google app's calendars to have any real use; Now however the powers that be finally discovered the smart re-scheduler and the other google tools for managing meetings and schedules; unfortunately I'm now in a position where I've got events notifications all over the place and because each of the calendars do not know about each other, I'm losing all of the advantages of rescheduler / free-busy status. TL;DR, 4 calendars, unified 'free busy' status without having to manually copy every event please.

  Read the article

• How do I stop GNU Freetalk from automatically filling in the buddy name?

  - by Journeyman Geek

  I'm using GNU Freetalk along with expect in order to send notifications to my phone – Freetalk has a readline interface, and I use expect to make a series of non interactive scripts that send information to another Jabber account. I'd like to have these scripts end freetalk 'properly' – that is to say user@domainname message one user@domainname message two /quit which would print out message one message two then quit. However Freetalk 'helpfully' adds user@domainame automatically so I get message one message two /quit as the output. The expect script still ends, but there's a delay. How would I stop Freetalk from adding in the 'buddy' address automatically?

  Read the article

• After upgrade to Windows 8.1, no Skydrive icon in task bar

  - by Jeremy Butler

  I upgraded to 8.1 and now I don't see the SkyDrive icon in the task bar, nor the green light under it indicating the my computer is communicating with SkyDrive. When I go into Control Panel |..|Notification Area Icons, SkyDrive is listed and noted as Up to date. However even though I mark it to Show Icons and Notifications I don't see any icon in the taskbar. Is it supposed to be this way, or do I have a problem? I kind of got used to that little green synchronizing indicator giving me a warm and fuzzy feeling.

  Read the article

• Constantly visible notification and access icon for Empathy in Gnome 3

  - by aef

  Since a short while I'm using Ubuntu Oneiric Ocelot (11.10) with gnome-shell (Gnome 3) and I'm trying to get accustomed to the default Empathy Instant Messaging client. One mayor problem for me (coming from Gnome 2 and Psi) is that there is no constantly visible icon which makes it clear (for example by changing its icon or showing an animation) if there are incoming messages which I did not read already and which lets me jump into them with one click. Also I'm missing a way to bring up the contact list or hide it away with a click. I sometimes have real problems even figuring out how to even open the contact list up again. Is there a Gnome 3 extension or some other trick available to display such a notifier in the top bar? I'm talking about something just like the sound and network controls which are already located there. I know that there are notifications in the lower notification area (former system tray), but as it is only visible as I move the mouse in the lower right corner of the screen, its useless for me.

  Read the article

• How do you pick what server setup you need?

  - by ed209

  I recently started receiving pubsub data feed from etsy. It averages around 250 notifications per minute. But obviously, when the USA wakes up that spikes quite heavily. I want to be able to deal with those spikes (about 3 per day) but the rest of day is fine. What's the best method of getting the right server configuration. My current approach is to keep upgrading until the server stops dying... next leap is: Processor: AMD Phenom II X6-1055T HEXA Core RAM: 4GB DDR2 SDRAM HD1: SATA Drive (7,200 rpm) (+500 GB 7200 RPM SATA hard drive) HD2: SATA Backup Drive (+500 GB SATA (7,200 rpm)) OS: Linux OS (+CentOS 5 64-bit) Bandwidth: 6000GB Monthly Transfer (3000 in + 3000 out) (+100M uplink port) What's the best approach to working out what sort of server setup you need?

  Read the article

• crontab still sending emails even with > /dev/null

  - by user2344668

  I have a crontab (root) that runs a script and output is set to /dev/null but I always get the emails whenever it runs. I only want to receive error emails. # Rackspace driveclient update (12pm MST) 0 12 * * * /root/scripts/driveclient-update > /dev/null The only way I can get it to turn off is to use /dev/null 2&1 but then I won't get error emails. This is happening on three different CentOS servers, two are 6.3 and one is 6.4. NOTE: I have read over and over that /dev/null is supposed to send stdout there and prevent the email if there is nothing but stdout from the script, so at works for at least some people; I cannot figure out why it is not working on these servers. Here's an example of where /dev/null is supposed to work: http://www.alphadevx.com/a/384-Suppressing-Cron-Job-Email-Notifications

  Read the article

• How to let Linux Python application handle termination on user logout correctly?

  - by tuxpoldo

  I have written a Linux GUI application in Python that needs to do some cleanup tasks before being terminated when the user logs out. Unfortunately it seems, that on logout, all applications are killed. I tried both to handle POSIX signals and DBUS notifications, but nothing worked. Any idea what I could have made wrong? On application startup I register some termination handlers: # create graceful shutdown mechanisms signal.signal(signal.SIGTERM, self.on_signal_term) self.bus = dbus.SessionBus() self.bus.call_on_disconnection(self.on_session_disconnect) When the user logs out, neither self.on_signal_term nor self.on_session_disconnect are called. The problem occurs in several scenarios: Ubuntu 14.04 with Unity, Debian Wheezy with Gnome. Full code: https://github.com/tuxpoldo/btsync-deb/tree/master/btsync-gui

  Read the article

• how to include error messages into backup reports for SQL Server 2008 R2?

  - by avs099

  Right now I have daily (differential) and weekly (full) backups set on my SQL Server 2008 R2 as jobs for SQL Server Agent with email notifications if job fails. I do get emails like this: JOB RUN: 'Daily backup.Diff backup' was run on 4/11/2012 at 3:00:00 AM DURATION: 0 hours, 0 minutes, 28 seconds STATUS: Failed MESSAGES: The job failed. The Job was invoked by Schedule 9 (Daily backup.Diff backup). The last step to run was step 1 (Diff backup). but often that happens because we delete/create new databases - and diff backup fails. And the only way for me to see the actual reason is to go to Log Viewer - Maintenance Plans logs. Is it possible to include "Error Message" field from the logs into notification emails? And more generic - is it possible to change notification email templates somehow?.. Thanks you.

  Read the article

• If email not received then do X (outlook 2013 on Exchange 2010)

  - by Brad

  I receive notification emails daily and would like to automate an easier way to manage all of those notifications. For example: Notification 1 from [email protected] is received daily between 10pm-1am Notification 2 from [email protected] is received daily between 12am-3am Notification 3 from [email protected] is received daily between 1am-4am I am looking for a way to page myself at [email protected] on my cellphone if any of these messages are not received within the defined time frame of when the email should have arrived. I would like to basically email a page like: ATTENTION Notification 2 not received within the allowed range. This way I would be notified instead of having to check the email manually and see that I only received 2 of the three alerts. Is there a way to do this in Outlook? Our exchange server is a hosted exchange server on GoDaddy if that info is needed.

  Read the article

• Question about server usage, big community platform [closed]

  - by Json

  Possible Duplicate: How do you do Load Testing and Capacity Planning for Web Sites I’m working on a community platform writen in PHP, MySQL. I have some questions about the server usage maybe someone can help me out. The community is based on JQuery with many ajax requests to update content. It makes 5 - 10 AJAX(Json, GET, POST) requests every 5 seconds, the requests fetch user data like user notifications and messages by doing mySQL queries. I wonder how a server will handle this when there are for more than 5000 users online. Then it will be 50.000 requests every 5 seconds, what kind of server you need to handle this? Or maybe even more, when there are 15.000 users online, 150.000 requests every 5 seconds. My webserver have the following specs. Xeon Quad 2048MB 5000GB traffic Will it be good enough, and for how many users? Anyone can help me out or know where to find such information, like make a calculation?

  Read the article

• Stop Metro on W8 from minimizing on dual screens

  - by Pestario

  I'm running Windows 8 with two screens, and have set it so that I have Metro on my secondary screen, because I thought that I could have that visible while I was working on other things on the main screen. The problem is, as soon as I click on something on the main screen, Metro goes away on my secondary screen. Is there any way to prevent this from happening?? I would like to be able to see f.ex the clock, notifications on mail/messaging etc. while I'm working on things.

  Read the article

• Is iTunes touching my library files automatically and/or in the background?

  - by Rich

  I am currently using a service called Cubby (it is like Dropbox) to sync my iTunes library between several computers. I would expect then that syncing only occurs when I add to or delete from the library. But I notice that often the Cubby app will be actively syncing when I haven't made any changes, and it will give notifications that (for example) a song changed which I definitely did not recently add/remove. Likewise I have lately been seeing weird changes in filename and track number tags. I wonder if this is iTunes trying to organize something automatically and failing? So does anyone know if iTunes will routinely modify your library files and how/why? What is it doing?

  Read the article

• Salesforce/PHP - Bulk Outbound message (SOAP), Time out issue - See update #2

  - by Phill Pafford

  Salesforce can send up to 100 requests inside 1 SOAP message. While sending this type of Bulk Ooutbound message request my PHP script finishes executing but SF fails to accept the ACK used to clear the message queue on the Salesforce side of things. Looking at the Outbound message log (monitoring) I see all the messages in a pending state with the Delivery Failure Reason "java.net.SocketTimeoutException: Read timed out". If my script has finished execution, why do I get this error? I have tried these methods to increase the execution time on my server as I have no access on the Salesforce side: set_time_limit(0); // in the script max_execution_time = 360 ; Maximum execution time of each script, in seconds max_input_time = 360 ; Maximum amount of time each script may spend parsing request data memory_limit = 32M ; Maximum amount of memory a script may consume I used the high settings just for testing. Any thoughts as to why this is failing the ACK delivery back to Salesforce? Here is some of the code: This is how I accept and send the ACK file for the imcoming SOAP request $data = 'php://input'; $content = file_get_contents($data); if($content) { respond('true'); } else { respond('false'); } The respond function function respond($tf) { $ACK = <<<ACK <?xml version = "1.0" encoding = "utf-8"?> <soapenv:Envelope xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/" xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <soapenv:Body> <notifications xmlns="http://soap.sforce.com/2005/09/outbound"> <Ack>$tf</Ack> </notifications> </soapenv:Body> </soapenv:Envelope> ACK; print trim($ACK); } These are in a generic script that I include into the script that uses the data for a specific workflow. I can process about 25 requests (That are in 1 SOAP response) but once I go over that I get the timeout error in the Salesforce queue. for 50 requests is usually takes my PHP script 86.77 seconds. Could it be Apache? PHP? I have also tested just accepting the 100 request SOAP response and just accepting and sending the ACK the queue clears out, so I know it's on my side of things. I show no errors in the apache log, the script runs fine. I did find some info on the Salesforce site but still no luck. Here is the link. Also I'm using the PHP Toolkit 11 (From Salesforce). Other forum with good SF help Thanks for any insight into this, --Phill UPDATE: If I receive the incoming message and print the response, should this happen first regardless if I do anything else after? Or does it wait for my process to finish and then print the response? UPDATE #2: okay I think I have the problem: PHP uses the single thread processing approach and will not send back the ACK file until the thread has completed it's processing. Is there a way to make this a mutli thread process? Thread #1 - accept the incoming SOAP request and send back the ACK Thread #2 - Process the SOAP request I know I could break it up into like a DB table or flat file, but is there a way to accomplish this without doing that? I'm going to try to close the socket after the ACK submission and continue the processing, cross my fingers it will work.

  Read the article

• Import Twitter XML into Visual Basic

  - by ct2k7

  Hello, I'm trying to import an XML that's provided by twitter into a readable format in Visual Basic. XML looks like: <?xml version="1.0" encoding="UTF-8" ?> - <statuses type="array"> - <status> <created_at>Mon Jan 18 20:41:19 +0000 2010</created_at> <id>111111111</id> <text>thattext</text> <source><a href="http://www.seesmic.com/" rel="nofollow">Seesmic</a></source> <truncated>false</truncated> <in_reply_to_status_id>7916479948</in_reply_to_status_id> <in_reply_to_user_id>90978206</in_reply_to_user_id> <favorited>false</favorited> <in_reply_to_screen_name>radonsystems</in_reply_to_screen_name> - <user> <id>20193170</id> <name>personname</name> <screen_name>screenname</screen_name> <location>loc</location> <description>desc</description> <profile_image_url>http://a3.twimg.com/profile_images/747012343/twitter_normal.png</profile_image_url> <url>myurl</url> <protected>false</protected> <followers_count>97</followers_count> <profile_background_color>ffffff</profile_background_color> <profile_text_color>333333</profile_text_color> <profile_link_color>0084B4</profile_link_color> <profile_sidebar_fill_color>ffffff</profile_sidebar_fill_color> <profile_sidebar_border_color>ababab</profile_sidebar_border_color> <friends_count>76</friends_count> <created_at>Thu Feb 05 21:54:24 +0000 2009</created_at> <favourites_count>1</favourites_count> <utc_offset>0</utc_offset> <time_zone>London</time_zone> <profile_background_image_url>http://a3.twimg.com/profile_background_images/76723999/754686.png</profile_background_image_url> <profile_background_tile>true</profile_background_tile> <notifications>false</notifications> <geo_enabled>true</geo_enabled> <verified>false</verified> <following>false</following> <statuses_count>782</statuses_count> <lang>en</lang> <contributors_enabled>false</contributors_enabled> </user> <geo /> <coordinates /> <place /> <contributors /> </status> </statuses> Now, I want to display it in a panel that automatically refresh after a certain period, however, I only want to pick out certain bits of info from that xml, such as profile_image_url and text and created_at. You can guess how the data will be formatted, much like that presented in TweetDeck and other Twitter clients. I'm quite new to Visual Basic, so how could I do this? Thanks

  Read the article

• SQL Server and Hyper-V Dynamic Memory Part 2

  - by SQLOS Team

  Part 1 of this series was an introduction and overview of Hyper-V Dynamic Memory. This part looks at SQL Server memory management and how the SQL engine responds to changing OS memory conditions.   Part 2: SQL Server Memory Management As with any Windows process, sqlserver.exe has a virtual address space (VAS) of 4GB on 32-bit and 8TB in 64-bit editions. Pages in its VAS are mapped to pages in physical memory when the memory is committed and referenced for the first time. The collection of VAS pages that have been recently referenced is known as the Working Set. How and when SQL Server allocates virtual memory and grows its working set depends on the memory model it uses. SQL Server supports three basic memory models:   1. Conventional Memory Model   The Conventional model is the default SQL Server memory model and has the following properties: - Dynamic - can grow or shrink its working set in response to load and external (operating system) memory conditions. - OS uses 4K pages – (not to be confused with SQL Server “pages” which are 8K regions of committed memory).- Pageable - Can be paged out to disk by the operating system.   2. Locked Page Model The locked page memory model is set when SQL Server is started with "Lock Pages in Memory" privilege*. It has the following characteristics: - Dynamic - can grow or shrink its working set in the same way as the Conventional model.- OS uses 4K pages - Non-Pageable – When memory is committed it is locked in memory, meaning that it will remain backed by physical memory and will not be paged out by the operating system. A common misconception is to interpret "locked" as non-dynamic. A SQL Server instance using the locked page memory model will grow and shrink (allocate memory and release memory) in response to changing workload and OS memory conditions in the same way as it does with the conventional model.   This is an important consideration when we look at Hyper-V Dynamic Memory – “locked” memory works perfectly well with “dynamic” memory.   * Note in “Denali” (Standard Edition and above), and in SQL 2008 R2 64-bit (Enterprise and above editions) the Lock Pages in Memory privilege is all that is required to set this model. In 2008 R2 64-Bit standard edition it also requires trace flag 845 to be set, in 2008 R2 32-bit editions it requires sp_configure 'awe enabled' 1.   3. Large Page Model The Large page model is set using trace flag 834 and potentially offers a small performance boost for systems that are configured with large pages. It is characterized by: - Static - memory is allocated at startup and does not change. - OS uses large (>2MB) pages - Non-Pageable The large page model is supported with Hyper-V Dynamic Memory (and Hyper-V also supports large pages), but you get no benefit from using Dynamic Memory with this model since SQL Server memory does not grow or shrink. The rest of this article will focus on the locked and conventional SQL Server memory models.   When does SQL Server grow? For “dynamic” configurations (Conventional and Locked memory models), the sqlservr.exe process grows – allocates and commits memory from the OS – in response to a workload. As much memory is allocated as is required to optimally run the query and buffer data for future queries, subject to limitations imposed by:   - SQL Server max server memory setting. If this configuration option is set, the buffer pool is not allowed to grow to more than this value. In SQL Server 2008 this value represents single page allocations, and in “Denali” it represents any size page allocations and also managed CLR procedure allocations.   - Memory signals from OS. The operating system sets a signal on memory resource notification objects to indicate whether it has memory available or whether it is low on available memory. If there is only 32MB free for every 4GB of memory a low memory signal is set, which continues until 64MB/4GB is free. If there is 96MB/4GB free the operating system sets a high memory signal. SQL Server only allocates memory when the high memory signal is set.   To summarize, for SQL Server to grow you need three conditions: a workload, max server memory setting higher than the current allocation, high memory signals from the OS.    When does SQL Server shrink caches? SQL Server as a rule does not like to return memory to the OS, but it will shrink its caches in response to memory pressure. Memory pressure can be divided into “internal” and “external”.   - External memory pressure occurs when the operating system is running low on memory and low memory signals are set. The SQL Server Resource Monitor checks for low memory signals approximately every 5 seconds and it will attempt to free memory until the signals stop.   To free memory SQL Server does the following: ·         Frees unused memory. ·         Notifies Memory Manager Clients to release memory o   Caches – Free unreferenced cache objects. o   Buffer pool - Based on oldest access times.   The freed memory is released back to the operating system. This process continues until the low memory resource notifications stop.    - Internal memory pressure occurs when the size of different caches and allocations increase but the SQL Server process needs to keep its total memory within a target value. For example if max server memory is set and certain caches are growing large, it will cause SQL to free memory for re-use internally, but not to release memory back to the OS. If you lower the value of max server memory you will generate internal memory pressure that will cause SQL to release memory back to the OS.    Memory pressure handling has not changed much since SQL 2005 and it was described in detail in a blog post by Slava Oks.   Note that SQL Server Express is an exception to the above behavior. Unlike other editions it does not assume it is the most important process running on the system but tries to be more “desktop” friendly. It will empty its working set after a period of inactivity.   How does SQL Server respond to changing OS memory?    In SQL Server 2005 support for Hot-Add memory was introduced. This feature, available in Enterprise and above editions, allows the server to make use of any extra physical memory that was added after SQL Server started. Being able to add physical memory when the system is running is limited to specialized hardware, but with the Hyper-V Dynamic Memory feature, when new memory is allocated to a guest virtual machine, it looks like hot-add physical memory to the guest. What this means is that thanks to the hot-add memory feature, SQL Server 2005 and higher can dynamically grow if more “physical” memory is granted to a guest VM by Hyper-V dynamic memory.   SQL Server checks OS memory every second and dynamically adjusts its “target” (based on available OS memory and max server memory) accordingly.   In “Denali” Standard Edition will also have sqlserver.exe support for hot-add memory when running virtualized (i.e. detecting and acting on Hyper-V Dynamic Memory allocations).   How does a SQL Server workload in a guest VM impact Hyper-V dynamic memory scheduling?   When a SQL workload causes the sqlserver.exe process to grow its working set, the Hyper-V memory scheduler will detect memory pressure in the guest VM and add memory to it. SQL Server will then detect the extra memory and grow according to workload demand. In our tests we have seen this feedback process cause a guest VM to grow quickly in response to SQL workload - we are still working on characterizing this ramp-up.    How does SQL Server respond when Hyper-V removes memory from a guest VM through ballooning?   If pressure from other VM's cause Hyper-V Dynamic Memory to take memory away from a VM through ballooning (allocating memory with a virtual device driver and returning it to the host OS), Windows Memory Manager will page out unlocked portions of memory and signal low resource notification events. When SQL Server detects these events it will shrink memory until the low memory notifications stop (see cache shrinking description above).    This raises another question. Can we make SQL Server release memory more readily and hence behave more "dynamically" without compromising performance? In certain circumstances where the application workload is predictable it may be possible to have a job which varies "max server memory" according to need, lowering it when the engine is inactive and raising it before a period of activity. This would have limited applicaability but it is something we're looking into.   What Memory Management changes are there in SQL Server “Denali”?   In SQL Server “Denali” (aka SQL11) the Memory Manager has been re-written to be more efficient. The main changes are summarized in this post. An important change with respect to Hyper-V Dynamic Memory support is that now the max server memory setting includes any size page allocations and managed CLR procedure allocations it now represents a closer approximation to total sqlserver.exe memory usage. This makes it easier to calculate a value for max server memory, which becomes important when configuring virtual machines to work well with Hyper-V Dynamic Memory Startup and Maximum RAM settings.   Another important change is no more AWE or hot-add support for 32-bit edition. This means if you're running a 32-bit edition of Denali you're limited to a 4GB address space and will not be able to take advantage of dynamically added OS memory that wasn't present when SQL Server started (though Hyper-V Dynamic Memory is still a supported configuration).   In part 3 we’ll develop some best practices for configuring and using SQL Server with Dynamic Memory. Originally posted at http://blogs.msdn.com/b/sqlosteam/

  Read the article

< Previous Page | 24 25 26 27 28 29 30 31 32 33 34 35  | Next Page >