Data on the network

Similar Questions

• Loss of data for the network flow

  I have an application where I need a chassis CRIO 9068 stream to a PC, and I'm running into some problems with data loss.

  The CRIO is sampling data, put in a picture 2D tanks (the size depends on the number of channels and the sampling frequency) and it streaming on the PC.  Data loss is not consistent, sometimes no data will be lost. Also the size of the array be listened to believe makes no difference, sometimes I won't lose data 24 channels 25 kech. / s, and sometimes I'll lose data 1-channel 10KS/s.

  The VI (host PC) runs in a non-timed while loop, it runs generally to about 50 Hz, sometimes hiccups system will make it slower for a loop or both.  Wirte CRIO loop runs at 20 Hz.  I think that the loss of data occurs when the PC has an if its hiccups, I know well that the size of the buffer to the stream network could absorb these hiccups.  If I make the size of the buffer is bigger, I seem to have a memory leak and the CRIO crashes after a few minutes.  I have also the CRIO record data directly without streaming, and no data is lost here, so I know its disappear somewhere in the stream.

  I also have a separate Standard TCP connection between for command and control, do not know if this could cause an interfereing with the network flow problem.

  I have attached photos of the screws where the endpoints are creadted and where data is written and read.

  I'm using Labview/RT 2013

  Thank you

  Suggestion of the AGJ is a good thing.  I had trouble keeping your code - there were so many things with wires everywhere...

  I have an application that runs on a PXI-8106 controller streaming 24 channels of data collected at 1 KHz (or 24K s/s).  He gets streaming via network to the host computer and from there flows streaming to disk.  At the same time, I'm streaming "Events" (transitions State, DIO line changes, etc.) to another stream from network to another data file.  A key step is that the data is off the network stream, they are put in a queue for sending the disk.  There are several parallel running loops, and I never lose a sample (I can tell because one of my channels is a "clock", the index of the loop timed running on the PXI system that produces the data that is sent to the host through the network stream - I can inspect this channel and see that each point is different from the previous one by 1).

  Do you use queues to stamp your data in and out of the network stream?

  BS

• Why convert? What is the advantage? Why convert it before you send the data over the network?

  Hello everyone.

  I have a question quite simple and intuitive. Yet Google does not seam to get the answer.

  Basically, I'm curious why should we catalogued something? It seems to me that it's just a little like a "band-aid" for editing data in conversion to a new data type, when it is not properly declared with an appropriate data type. What would be the advantage of this?

  Furthermore, I see a lot of data type casting to U8 programs before sending it over the network. Various reviews say it's for speed, but I can't understand how it improves speed. The same amount of data is still sent, just in a different type of data. This is overcoming a limitation of the network. The network somehow send smaller data types faster?

  I've been curious about this for a while and have yet to find an answer. Any help is appreciated.

  Thank you
  Taylor S. Amarel

  tsa256 wrote:

  Basically, I'm curious why should we catalogued something? It seems to me that it's just a little like a "band-aid" for editing data in conversion to a new data type, when it is not properly declared with an appropriate data type. What would be the advantage of this?

  It really depends on the specific situations.

  For example, some functions explicitly require certain types of data as inputs if they are not polymorphic.

  This means that if you do not convert the data yourself and you will see a red dot forced LabVIEW will do it for you.

  As to why the data was not the correct data type in the first place, it is sometimes inevitable.

  A few examples

  -What happens if you just read a text file as a long string.

  If there is digital data in this file, you need to analyze, and then convert it to a usable type (int, sgl, dbl, etc.)

  -you are given a 3rd part VI, dll, etc. that returns a numeric value as a double, but you need the value to I8.

  You cannot change the output of the function given to you, so what you can do to remedy this situation is cataloged data.

  tsa256 wrote:

  Furthermore, I see a lot of data type casting to U8 programs before sending it over the network. Various reviews say it's for speed, but I can't understand how it improves speed. The same amount of data is still sent, just in a different type of data. This is overcoming a limitation of the network. The network somehow send smaller data types faster?

  Somehow, this is my first response, but in a more general sense.

  Many functions down to the level of the OS run in binary.

  If your data are not already in binary, the sending computer must convert it to binary, and then send it, have the receiving computer convert in return.

  Two conversions on each side of the transmission add unnecessary delays.

• Import data over the network link in oracle 11 g

  We want to take the export of the NDO schema in database production and
  import in the scheme of the NDO in UAT database on a network
  link using data in Oracle 11 g pump. Kindly share the steps.

  Scenario:
  Directly import the schema of TEST01 in the production database (oraodrmu) to test the oraodrmt database, on
  a network to help pump link and data from database in Oracle 11 g.

  Note: When you perform an import via a database link, the import source is a database, not a dump file set, and the data is imported to the connected instance of database.
  Because the link may identify a database remotely in a network, the database of terms link and the network link are used interchangeably.

  =================================================================
  STEP 1 (IN PRODUCTION DATABASE - oraodrmu)
  =================================================================

  [root@szoddb01] > su - oraodrmu

  Enter the user name: virtue sysdba
  Connected to:
  Oracle Database 11 g Enterprise Edition Release 11.2.0.2.0 - 64 bit Production
  With partitioning, OLAP, Data Mining and Real Application Testing options

  SQL > grant resources to test01.

  Grant succeeded.

  SQL > grant imp_full_database to test01.

  Grant succeeded.

  SQL > select owner, object_type, status, count (*) in the dba_objects where owner = "TEST01" group by owner, object_type, status;

  OWNER OBJECT_TYPE STATUS COUNT (*)
  ------------------------------ ------------------- ------- ----------
  TEST01 PROCEDURE VALID 2
  TEST01 TABLE 419 VALID
  TEST01 SEQUENCE VALID 3
  TEST01 FUNCTION VALID 8
  TEST01 TRIGGER 3 VALID
  TEST01 INDEX VALID 545
  TEST01 LOB 18 VALID

  7 selected lines.

  SQL >
  SQL > define pages 999
  SQL > format 'size Mo' 999 999 999 col
  SQL > col 'objects' format 999 999 999
  SQL > select obj.owner 'owner '.
  obj_cnt 2, 'objects '.
  3, decode (seg_size, NULL, 0, seg_size) 'size MB.
  4 of (select master, the obj_cnt of count (*) group by owner dba_objects) obj
  5, (select master, ceil (sum (bytes) / 1024/1024) seg_size)
  6 Group dba_segments by owner) seg
  7 where obj.owner = seg.owner (+)
  Order of 8 by 3 desc, desc 2, 1
  9.

  Size of the objects owner MB
  ------------------------------ ------------ ------------
  NDO 8 097 284 011
  9 601 1 912 SYS
  998 1 164 TEST01

  3 selected lines.

  SQL > exit

  =================================================================
  STEP-2 (in TEST DATABASE - oraodrmt)
  =================================================================

  [root@szoddb01] > su - oraodrmt

  [oraodrmt@szoddb01] > sqlplus

  SQL * more: Production release 11.2.0.2.0 on Mon 3 dec 18:40:16 2012

  Copyright (c) 1982, 2010, Oracle. All rights reserved.

  Enter the user name: virtue sysdba

  Connected to:
  Oracle Database 11 g Enterprise Edition Release 11.2.0.2.0 - 64 bit Production
  With partitioning, OLAP, Data Mining and Real Application Testing options

  SQL > select name, open_mode from v$ database;

  NAME OPEN_MODE
  --------- --------------------
  ODRMT READ WRITE

  SQL > create tablespace test_test datafile ' / trn_u04/oradata/odrmt/test01.dbf' size 2048 m;

  Created tablespace.

  SQL > create user identified by test123, default tablespace test01 test_test;

  Created by the user.

  SQL > grant resources, create session to test01.

  Grant succeeded.

  SQL > grant permission to test01.

  Grant succeeded.

  SQL > grant imp_FULL_DATABASE to test01.

  Grant succeeded.

  Note: ODRMU is the name of hoste DNS. We can test the connection with: [oraodrmt@szoddb01] > sqlplus test01/test01@odrmu

  SQL > create directory test_network_dump as ' / dbdump/test_exp ";

  Created directory.

  SQL > grant read, write on directory test_network_dump to test01.

  Grant succeeded.

  SQL > conn test01/test123
  Connected.

  SQL > DATABASE LINK remote_test CONNECT TO test01 identified by test01 to create with the HELP of "ODRMU";

  Database link created.

  To test the link of database, we can try the sql below:

  SQL > select count (*) in OA_APVARIABLENAME@remote_test;

  COUNT (*)
  ----------
  59

  SQL > exit

  [oraodrmt@szoddb01] > impdp test01/test123 network_link = remote_test = test_network_dump remap_schema = test01:test01 logfile directory = impdp__networklink_grms.log;
  [oraodrmt@szoddb01] >

  Import: Release 11.2.0.2.0 - Production on Mon 3 19:42:47 dec 2012

  Copyright (c) 1982, 2009, Oracle and/or its affiliates. All rights reserved.

  Connected to: Oracle Database 11 g Enterprise Edition Release 11.2.0.2.0 - 64 bit Production
  With partitioning, OLAP, Data Mining and Real Application Testing options
  Departure "TEST01". "' SYS_IMPORT_SCHEMA_01 ': test01 / * network_link = remote_test = test_network_dump remap_schema = test01:test01 impdp_grms_networklink.log = logfile directory
  Current estimation using BLOCKS method...
  Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
  Total estimation using BLOCKS method: 318,5 MB
  Processing object type SCHEMA_EXPORT/USER
  ORA-31684: USER object Type: "TEST01" already exists
  Processing object type SCHEMA_EXPORT/SYSTEM_GRANT
  Processing object type SCHEMA_EXPORT/ROLE_GRANT
  Processing object type SCHEMA_EXPORT/DEFAULT_ROLE
  Processing object type SCHEMA_EXPORT/PRE_SCHEMA/PROCACT_SCHEMA
  Object type SCHEMA_EXPORT/SEQUENCE/SEQUENCE of treatment
  Object type SCHEMA_EXPORT/TABLE/TABLE processing
  . . imported "TEST01". "' SY_TASK_HISTORY ' 779914 lines
  . . imported "TEST01". "' JCR_JNL_JOURNAL ' 603 lines
  . . imported "TEST01". "" 1229 GX_GROUP_SHELL lines.
  . . . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .
  . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..
  . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..
  . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..

  Work "TEST01". "" SYS_IMPORT_SCHEMA_01 "completed with error (s 1) at 19:45:19

  [oraodrmt@szoddb01] > sqlplus

  SQL * more: Production release 11.2.0.2.0 on Mon 3 dec 19:46:04 2012

  Copyright (c) 1982, 2010, Oracle. All rights reserved.

  Enter the user name: virtue sysdba

  Connected to:
  Oracle Database 11 g Enterprise Edition Release 11.2.0.2.0 - 64 bit Production
  With partitioning, OLAP, Data Mining and Real Application Testing options

  SQL > select owner, object_type, status, count (*) in the dba_objects where owner = "TEST01" group by owner, object_type, status;

  OWNER OBJECT_TYPE STATUS COUNT (*)
  ------------------------------ ------------------- ------- ----------
  TEST01 PROCEDURE VALID 2
  TEST01 TABLE 419 VALID
  TEST01 SEQUENCE VALID 3
  TEST01 FUNCTION VALID 8
  TEST01 TRIGGER 3 VALID
  TEST01 INDEX VALID 545
  TEST01 LOB 18 VALID
  TEST01 DATABASE LINK VALID 1

  8 selected lines.

  SQL >
  SQL > define pages 999
  SQL > format 'size Mo' 999 999 999 col
  SQL > col 'objects' format 999 999 999
  SQL > select obj.owner 'owner '.
  obj_cnt 2, 'objects '.
  3, decode (seg_size, NULL, 0, seg_size) 'size MB.
  4 of (select master, the obj_cnt of count (*) group by owner dba_objects) obj
  5, (select master, ceil (sum (bytes) / 1024/1024) seg_size)
  6 Group dba_segments by owner) seg
  7 where obj.owner = seg.owner (+)
  Order of 8 by 3 desc, desc 2, 1
  9.

  Size of the objects owner MB
  ------------------------------ ------------ ------------
  NDO 8 065 247 529
  9 554 6 507 SYS
  999 1 164 TEST01

  13 selected lines.

  =================================================================
  STEP 3 TO REMOVE THE LINK OF DATABASE
  =================================================================

  [oraodrmt@szoddb01] > sqlplus

  SQL * more: Production release 11.2.0.2.0 on Mon 3 dec 19:16:01, 2012

  Copyright (c) 1982, 2010, Oracle. All rights reserved.

  Enter the user name: virtue sysdba

  Connected to:
  Oracle Database 11 g Enterprise Edition Release 11.2.0.2.0 - 64 bit Production
  With partitioning, OLAP, Data Mining and Real Application Testing options

  SQL > drop database link remote_test;

  Database link dropped.

• VCD api to collect data on the network

  Hi all

  Anyone know the api used to get the external ip address of a network of VAPP is nat translated to a network of org.

  see you soon,

  John

  just in case anyone needs it in the future I found the answer with

  networkConnectionSection

  or networkConnection

• Dreamweaver CC 2105 "internal data error - failed to receive data from the network.

  I haven't seen this error before and have used Dreamweaver for many years. But, I am getting very regularly (it's more than intermittent) since the last update of Dreameaver. I'm just checking if it's a bug in Dreamweaver, or a bug of Yahoo Web Hosting.

  What happens is that I log on my SFTP site (sometimes I can not yet connect), then try to download a file and get the message "an FTP error occurred - cannot establish a connection to the host." Internal data error. Unable to connect to the server. ».

  However, and this is the really, really bad bits, the file that WAS on the server (i.e. who was to be replaced) seems to be DELETED, i.e. * before * the update is applied (which is not because of the error). I don't know if Dreamweaver somehow did this or not, but this means that if you were, say, download a revision to top-level index.html file in your site, this file is no longer there, and visitors to your site then sees the * entire * structure of the files in your site - potentially catastrophic.

  I tried to empty the cache to Dreamweaver (http://www.dmxzone.com/go/16740/clearing-dreamweaver-s-cache/), but that makes no difference.

  Did someone else presents itself has found a solution for it (or identified the problem)?

  Thanks in advance!

  There was a few other messages on Yahoo and connection problems recently.

  Have you tried downloading using another FTP, Filezilla client program, yet?

  If the problem that is happening, it may just be a problem with Yahoo. If it isn't, at least your site will be updated while troubleshooting continues with DW.

• Get the latest data of Variable shared before disconnecting from the network

  Hello.

  I use Shared - published network variables for my communication between two PC's. Each LV software will pass the data to another.

  Sometimes the wireless connection to become really bad and the network connection stops.

  I want to get the latest data. Currently my VI will show the value '0' on the variable shared data once the connection is cut.

  I tried using shift, but she registers the same results.

  Can someone show me how I can stay past data when the network connection is disconnected.

  Thank you very much.

  Kind regards

  Hello Fan, Ravens

  It works! Thank you very much.

  Kind regards

• Using too much CPU only when loading the network

  When FF works closely with the network (wired or wireless) card for access to the network, CPU utilization spikes in way more than 70% and he stayed there until the page loading process finishes web.
  After the upload is complete, the CPU usage remains somewhere between 5-10%.

  When I look at clips youtube 1080 p, where FF is duty next buffer part of this clip I watch, new CPU use peaks above 70% with the flash plugin and gives a total of a 100% of cpu for a few moments and use which make this clip to stop playing or to frame bad times, and ties.

  I have been activated in about: config webgl acceleration to activate the force and that much improve the initial problem, then I activated gfx.direct2D - activated force and also to improve the problem remaining.
  I also disable the safe_mode parameter file adobe flash config protectedmode = 0.

  I have to mention that the initial problem was way to bad ... 
  

  Now I get to point when I don't know what to do so that FF does not use much CPU loading pages.
  As you can see I use to publish this version of Nightly 64 bit (I think it's the 64-bit version...) which is much more better and I would like to read youtube 1080 p clip normally, after that I have wait a few moments to buffer, before starting to play.
  Version 32 bit FF won't let me watch YouTube without framing hurt when the clip is buffered.

  I was monitoring the Task Manager and all the processes listed... and there, I can't blame antivirus or firewall to the peaks of the processor. Also if I do the rest of the network traffic, as well as inside the network custom CPU even file copy flinch.

  Can someone help or give an opinion on what setting to do, so I can eliminate these spikes?

  Thanks in advance!

  Problem solved partially:

  I have observed that the CPU spikes occur not only while I use FF or mozilla nightly, but on all web browsers (Chrome, IE, Opera, etc) that I have used, while these browsers accessing the network.

  Uninstalling the firewall of the solution of the AV portion has helped a lot and by also disabling the bed in the firewall solution of W7, but for long term use, this isn't the right answer (for my at least). Without any firewall FF work or installed works like a charm, while the CPU usage is increased at a decent level of 50% max (total tasks and running services)!

  Anyway, as I said earlier, my pc without any firewall operation is not a good think to do, so I'll replace my computer with a faster, or I'm going to overclock it at a level that zapping INTERNET and navigation will not infringe too much on the performance of the system, or... I live whit it and I accept the limitations...

  Perhaps, in the development of the FF, programmers will keep in mind that not everyone on this planet has powerful processors - and... give a few options... to decrease the CPU at acceptable levels, event, even if they use the firewall solutions that violate on the CPU while FF is buffering of data across the network.

  I should also mention that Mozilla nightly (64-bit) and FF, are always better then IE - in my opinion - which is.

• Measure the impact of the network stack / components in latency

  Hello

  If I want to measure the impact of VSAN network as part of the latency, how better would be to do?

  Example: Writing takes place. It is of course delivered on VSAN network to another host as a component of the object is located here also. Transport data on the network and putting to the physical disk layer take time.

  VSAN using observer, it is easy to see the latency time physical disks and how much they contribute to the latency. Looking latency compared to the level of 'customer' shows, however, a different image which is logical because it is high in the stack.

  As I can not afford for 10 Gbit Ethernet, my Lab network is a 1 Gbps where each node has two NIC in a LACP LAG, and I see clearly a bow to talk to another node in the network more than 1 NETWORK and talk to another node on the other card NETWORK LACP nicely splits the traffic on the two cards.

  During reconstructions, I noticed that a nice full use of those two links on servers, so from a bandwidth perspective, using LACP as seems to be worth. But it would increase the latency to an undesirable extent? I have also noticed that during normal operations, the links never 1Gig not even approaching is saturated, sort of presumption 10 giga Ethernet unjustified for my small environment 4 node (i 80VM on this our).

  I want to be able to do is determine how much latency is added by the network layer.

  Hi John,.

  I threw out the 850 Pro SATA Samsung replace them with HCL - Listed SAS SSD. The difference is simply huge. But I still have weird 100ms + 'moments of despair' sometimes.

  I use vDS with a single Interface with LACP vmKernel on a pair of chassis-Stack switches Cisco Gigabit of SG500, where the LACP trunk is spread over the two chassis. When you look at observer vSAN during these 'unhappy moments' where latency is very high from the point of view of the virtual machine, I see that in terms of 'disc', the latency is extremely low (this used to be very high with the Samsungs) but in the levels 'Owner' and 'Customer', the latency can be very high. I suspect it's due to the latency of the network.

  I will dissolve the LACP, configure a second interface with a different subnet vmKernel and prevent traffic from the cross-frame. In this way, a lot of overhead should be gone. I'll post the results.

  My logical vSAN observer troubleshooting is:

  If

  VM feel morose

  and

  the metric "virtual disk" in vCenter is high (explaining why the virtual machine feels slow when participates in the activity of the disc)

  While

  Use memory and CPU of the host are low

  and

  latency to the 'disc' level is very low

  but

  latency on the 'owner' and 'client' levels are high

  then

  blame the network

  endif

  Does this make any sense?

  Find out what network component is the culprit is another question...

• How to download all Win7 CS6 Prod. Prem updates for the network?

  How to download all Win7 CS6 Prod. Updates bonus once to install on network deployment workstation?

  Updates are 1.8 GB and need to download once for deployment, is not up to date on the network for machines...

  Post edited by: STJAtechguys

  You can download the updates manually for all applications of Prod Prem CS6, from location below.

  http://www.Adobe.com/downloads/updates/

• Cellular data opens, but may not use the network have to reset net work every time

  Hello everyone

  My problem is:

  When you use the cell data, after several hours, the net is not available

  I have to reset the network every time

  and this happens every day

  I'm so sad!

  Hello

  If you have not already done so, it can help to upgrade your iPhone to the latest version of the software:

  • Update the software on your iPhone, iPad or iPod touch - Apple Support iOS

  Also, follow the steps detailed here:

  • If you cannot connect to a cellular network or cellular data - Apple Support

• 63 error to TCP connection open in Simple data Client.vi when connecting to the network storage device.

  Hello

  I tried the simple data client and simple example of data server. I tried this example to connect to a NAS device. My laptop properly communicates with the device since I ping the address of the NAS device and discovered it works perfectly fine. Also, I put the static IP of my pc even as the NAS. So everything I've done. But when I run the program giving the ip address of a NAS device, it shows an error 63.

  Error 63 to TCP connection open in Simple data Client.vi

  Possible reasons: LabVIEW: serial port receive buffer overflow. = LabVIEW: the network connection was refused by the server.

  Why is happening. What can be done to solve this problem. Thanking you.

  vindsan wrote:

  What is the firewall of the SIN.

  Yes, it is very likely. Try disabling the firewall and try again. If the code works when you run the Server & client on your machine, but not when you run the server on your NAS and client on your machine (assuming you have entered the correct port for the NAS server), then it is very likely to be a firewall problem.

  vindsan wrote:

  Server that is listening is no longer VI also means what.

  Your VI server listens only once for a connection after receiving an incoming connection (or it expires after 60 years), he stops listening and the port will be more open for incoming connections - you must re - run the VI for another connection. With your code, you must run the VI server and then run the VI client in the 1960s, so it's not expire.

• How to create a Subscriber variable to the network & OnDataUpdate for Variant data type?

  Looking at the example C:\Documents and Settings\All Users\Documents\National Instruments\MStudioVS2008\DotNET\Examples\NetworkVariable\Basic\cs\Subcriber.2008, how can I create a subscriber to the network variable and OnDataUpdate for Variant data types?

  All tracks will be useful. Thank you.

  I solved it by using the correct data type in .NET, for example an sbyte to CVIInt8.

• Limit the use of data on home network

  Hello

  I currently have three computers connected to a wifi modem with two computers running vista and a Mac SL running. I was wondering if it was possible to limit the daily use of computer data? I have experimented with the settings of the router/modem, but there is no built-in feature.

  Modem: 2701 HGV - W (bigpond)

  Any help would be fantastic.

  I have no idea on the SL Mac (you will need to post this question in a Mac forum or contact support Mac), but as far as Vista goes, certain limits can be defined in part by using parental controls and in part by using permissions (but it depends on exactly what you want to say that I am addressing the following).  No process will actually limit the use of network data (although they can help you to set other boundaries that can help).

  In Vista, here is the procedure for setting up restrictions. http://windows.microsoft.com/en-US/windows-vista/Set-up-Parental-Controls. Once you have access to parental controls, and then go to the section of deadlines and follow the instructions there (or for the other sections).  Technically, it's not a way to limit the use of data in itself - if by here you mean the amount of data accessed for a specified period (but you can limit the data that is used, what programs are available, what websites are allowed, when the computer can be used and several other things.  To limit the available data, you use permissions rather than Parental control, but well enough that is already treated as system implements security rights to prevent users other data and you can do this for any folder that you create outside profiles (while the file system are almost already protected - even on your part).  If you mean data over the Internet, then there is no way I know in Vista or IE which limit explicitly by the user.  Keep in mind that this will only work with standard user accounts as account administrator can simply replace the limitations or remove them all.  I know that's not what you asked really, but I thought that I should bring him as he is very much related.

  Now for the good news that you want can be done with the 3rd party software free.  See the following:http://www.ampercent.com/control-internet-data-usage/4572/.  This seems to be exactly what you want and should do the trick.

  I hope this helps.

  Good luck!

  Lorien - MCSA/MCSE/network + / has + - if this post solves your problem, please click the 'Mark as answer' or 'Useful' button at the top of this message. Marking a post as answer, or relatively useful, you help others find the answer more quickly.

• Deliver the addition of new set of switches stacked as an extension to the network in another data center building

  Hello

  I was wondering if someone can help me, I am a humble sysadmin with no real network admin training.

  I have the following Setup to work in a rack in our data center.

  2 x Dell Sonicwall NSA2400
  2 x stacked switches Dell Powerconnect N5524

  A Sonicwall is connected to each switch (active-passive)

  Every Dell server has 2 x nics in servile failover mode, one of each network card connected to each switch

  This is done to allow a full redundancy for network, cable, switches and Firewall interface cards. It works well.

  The above mentioned rack is now full and we ordered a new rack that is in a different datacenter of construction on the same site. We decided to buy another set of stacked switches and _extend_ (no battery) these via optical SPF on the rack on the other.

  The idea is that we still have the switch fault tolerance but link to two grids through 10 GB. The new switches are 2 x Dell Powerconnect N2048 and have been implemented stacked with any connected devices.

  I have implemented SPF ports on both switches to be a junction ports and connected the two sets of switches, unfortunately our site went offline, our alert went ballistic and no servers could not talk to each other!

  Quickly, I unplugged the new switches and after a worrisome 10 alerts deleted seconds and the site is back online.

  I don't know why this has happened and reflected on the STP parameters that my googleing seem well. I don't know I'm missing something basic.

  I'll post a framework and a diagram in a follow-up post

  Which is no doubt tree covering the shot weight. It is a good idea to assign lower priority to the main switch. Generally, your root switch will be the one that plugs into the firewall. Definition of priority on the main switch to 4096, it will make the root switch. So as you work your way through different layers of the network, increase the priority. Access layer switches can be left by default.

  I would also check the physical connections for loops. In situations where you have multiple connections switch, make sure that these ports are configured in a SHIFT.

  Keep us informed.