Admin Command Timeout

My streaming server 3 hanged last night and can be restarted.
I found that the admin logs have thousands of line like the
following:
2008-12-22 03:20:15 236 (e)2671287 Admin Command Timeout:
(Command (getServerStats) timed out.) -
2008-12-22 03:20:15 236 (e)2671287 Admin Command Timeout:
(Command (getVHostStats) timed out.) -
How can I solve this problem? Restart the computer helps, but
I wonder there is a way to solve it.

Well, really, none of these problems should happen! But,
sadly, they do and not just for me.
A Google search will show that these Admin timeouts are a
very common problem but no-one seems to have a solution. More
worryingly, I have never seen any suggestion from Adobe about what
might cause them.
Since my original post I have made two changes to our app:
(1) when the 'owner' of the room leaves I unload the instance
(after a 30 seconds delay)
(2) I amended the application.xml for this app to timeout
idle connections after 2 minutes (the default is no timeout)
This has made things slightly better in that the app now runs
for 2/3 days without failing (rather than failing twice a day).
But, it still fails and the only thing that brings it back is a
reboot of the server which is totally unacceptable in a live
environment where we usually have 500+ clients watching 100 live
streams at any one time (many on a pay-per-minute basis). The only
other significant thing I notice now is that once the machine has
been running for a couple of days the logging stops completely for
some (but not all) instances i.e. nothing is displayed in the admin
console and nothing is written to the log files. It's as if the FMS
Admin has just given up.
Re. your suggestion about checking IPs etc. There are no
"hidden clients"! When the Admin fails it is possible to be on a
machine with a single connection (i.e. my client). When I reload my
client there is a trace that I have disconnected, then reconnected
again *but* the connection count goes up by one. If I reload twenty
times the count will increase to 20. Even if I close my browser the
connections will apparently stay, sometimes for days!
As I mentioned, this happens on multiple servers with lots of
different clients and very, very low load. When it happens is, so
far, unpredictable other than it will happen and bring down our
live system.

Similar Messages

  • Admin Commands issue on Windows Web Server 2008 R2 - 64Bit

    Hi,
    I have tried exactly same process on FMS3.5 Installed on Centos and it worked perfectly.
    But not working in case of Windows Web Server 2008 R2 - 64Bit.
    Here is what i am trying to do -
    Basically i want to extract live streams list and display them in HTML as ordered list on another server them FMS.
    What i did in the case of FMS on Centos server -
    I installed Apache and PHP on this server and created a PHP file with following text in it. -
    abc.php
       $xml = simplexml_load_file('http://<fms-server-ip>:1111/admin/getLiveStreams?auser=admin&apswd=password&appInst=live','Simple XMLElement', LIBXML_NOCDATA);
    foreach($xml->children() as $child)
       foreach($child->children() as $child2)
              echo $child2 . ",";
    and in hosting server where i want to display the list i entered following code in PHP file -
    $homepage = file_get_contents('http://<fms-server-ip>/abc.php');
    echo $homepage;
    i got comma seperated values easily and everything is loading fast without any issue.
    Now problem occured with Windows Web Server 2008 R2 - 64Bit.
    I installed FMS, Apache and PHP in this server also, did evrything in the same way like in centos.
    FMS, APache and PHP files are working fine in this.
    But only problem occured when in abc.php i use
    $xml =  simplexml_load_file('http://<fms-server-ip>:1111/admin/getLiveStreams?auser=admin&apswd=password&appInst=live','Simple XMLElement',  LIBXML_NOCDATA);
    After this file do not loads and takes longer time to load and after some time a blank page appears. It works fine with a local XML file.
    i am able to see the XML file by opening it directly in browser, but somehow this is not working with   simplexml_load_file. I dont think this is PHP issue because with local XML file it works fine. May be some port issue or something else. But i am able to view XML file directly in browser from another computers as well.
    Kindly help, is there some other configurations should be changed in case of windows version. I have allowed admin commands as usual before doing all this. Ports 1111,80,1935 are also opened.
    Please guide

    Hi,
    Yes in Browser, the response is perfect. The only issue is in Calling inside in any file, i even tried Javascript's XML dom to load it, but same result, the file is just loading and never loads completely. I also tried using an SWF file with AS3 code to load XML inside as3 and send the result to any JS function.
    The output comes using trace in as3, but when it comes to browser, the same loading happends and result never loads.

  • Creating a Global Role using weblogic.Admin command

    Hi,
    Does anyone have an example of creating a global role using the weblogic.Admin commands? I think I have to use the INVOKE command with the DefaultRoleMapper and createRole method, but I'm not quite sure what the rest of the syntax is.
    Thanks,
    Gabriel

    Gabriel,
    The following works for me:
    weblogic.Admin -url t3://localhost:80 -username weblogic -password weblogic INVOKE -mbean "Security:Name=myrealmDefaultRoleMapper" -method createRole "" "MyGlobalRole" "Grp(Administrators)" ""
    The null first parameter identifies this role as a global role.
    The second param is the name of the role.
    The third parameter is the policy expression. Here, I've mapped the role to the Administrators group. You can also map it to users or a combo of the two. For example, to map it to the "weblogic" user, use "Usr(weblogic)" as the policy expression. If you leave this parameter empty, the role will be created but will not be mapped to anything.
    I'm not sure what the fourth parameter is for. It's not defined in the RoleEditorMBean docs but not including it causes an error. I suspect it's a description field because WLS does not seem to care what you put there.
    HTH,
    Mike

  • How to set command timeout for table object?

    I have a report that is generating an error during rendering in a CrystalReportViewer control (v10.5).  The error is "Failed to open a rowset." which appears to be caused by a timeout.  I'm currenlty setting the connection info and location of the sproc programmatically.  How can I set the command timeout as well?  Can I set this via the Attributes collection of the ConnectionInfo object?  (This appears to be a command timeout, not a connection timeout, which I'm already including in the Attributes collection of the ConnectionInfo object and setting at 15 seconds.  I've tried making this as much as 60 seconds but it doesn't change the behavior.)

    The error message includes text that states the timeout was exceeded (in bold below)...
    CrystalDecisions.CrystalReports.Engine.InternalException: Failed to open a rowset.
    Details: ADO Error Code: 0x
    Source: Microsoft OLE DB Provider for SQL Server
    Description: Timeout expired
    SQL State: HYT00Error in File J:\DOCUME1\ADMINI1\LOCALS~1\Temp\SpecificationSummaryReport {42EDFEA3-DD0D-4F5E-BC0B-F9CC1126A872}.rpt:
    Failed to open a rowset. ---> System.Runtime.InteropServices.COMException (0x800002D3): Failed to open a rowset.
    Details: ADO Error Code: 0x
    Source: Microsoft OLE DB Provider for SQL Server
    Description: Timeout expired
    SQL State: HYT00Error in File J:\DOCUME1\ADMINI1\LOCALS~1\Temp\SpecificationSummaryReport {42EDFEA3-DD0D-4F5E-BC0B-F9CC1126A872}.rpt:
    Failed to open a rowset.
    1)  Is there a Service Pack for CR 10.5???  That is the runtime that is in use - not XI.
    2)  The report viewer app is in version 10.5 (as mentioned in the first post) but the report was created in the designer for version 11.  Before we assume that is a problem - I have other reports created this way that work fine.
    3)  The data source is SQL 2005.
    4)  The connection type is OLE DB - Microsoft OLE DB provider for SQL Server.
    I think the issue is due to corruption in the report file.  I've seen this many times with reports (over the last 10 years).  I've had reports that refuse to work but if I recreate them from scratch the new report works fine.  In fact, I've recreated the data portion of this report (using the same sprocs) and that one works.

  • Set Command Timeout For SQL Server

    Hello,
    How do you set the command timeout for the 'open' statement? I am running a Database via SQL Server and I tried the dialog box timeout statement (the check box on the advanced tab) and simply get an error. The manual shows a 'step.commandtimeout' but how is it implemented?
    Thanks,
    Kevin

    Kevin -
    I looked at the internal implementation of the CVI SQL Toolkit. The toolkit function that the step type uses is DBNewSQLStatement. Internally to the toolkit the function opens a recordset instead of a command object. The command timeout attribute is only available on a command object and not on a recordset object, so as implemented inside the toolkit the error is appropriate. I think the toolkit could have used a command object to create the recordset object and this would allow the toolkit to let you set the timeout attribute. Unfortuneately this is not the way it was done.
    For the future I may investigate to see if there is a way to bypass this limitation by using different toolkit functions, but I am not sure if there a
    re any side effects.
    Scott Richardson (NI)
    Scott Richardson
    National Instruments

  • ATA command timeouts after disk has gone sleeping

    Hi all
    I've configured myself a NAS, which is infrequently accessed, so I set the standby timer of the disks to 241 (30 minutes) using hdparm.
    for i in /dev/sd?; do
        hdparm -S 241 $i > /dev/null
    done
    After 30 minutes the disks go to sleep, no problem there, but once they are accessed again, the first command sent to the disk time out:
    ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
    ata7.00: cmd b0/d8:00:00:4f:c2/00:00:00:00:00/00 tag 0
             res 40/00:00:00:f1:00/00:00:00:00:00/00 Emask 0x4 (timeout)
    ata7.00: status: { DRDY }
    ata7: hard resetting link
    ata7: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
    ata7.00: max_sectors limited to 256 for NCQ
    ata7.00: max_sectors limited to 256 for NCQ
    ata7.00: configured for UDMA/133
    sd 6:0:0:0: timing out command, waited 7s
    ata7: EH complete
    SCSI device sdg: 1953525168 512-byte hdwr sectors (1000205 MB)
    sdg: Write Protect is off
    sdg: Mode Sense: 00 3a 00 00
    SCSI device sdg: drive cache: write through
    After that the drives work fine again. The command timeout is 7 seconds, which seems too low because the disks need some time to spin up.
    Two questions:
    Can the ATA command timeout be increased?
    If the command fails, is it retried?
    Thanks,
    Glenn

    The G5 Discussion is here while this is devoted to Mac Pro, hence the confusion. The MP doesn't use cables, just trays that slide in.
    Is it possible you have backups you can boot from? (hint, hint). Or Disk Warrior 4? would also be 'nice.'
    Hard restarts DO trash files and directories. And never make changes w/o a bootable backup. Maybe a couple FW drives. (One person yesterday backed up his problem G5 drive only that turned his 'good backup' into an unbootable drive also, and didn't have a 2nd backup - which can happen and why you don't want to rely on just one).

  • Does Windows 8.1 support NVMe admin commands to access SMART info from SSD?

    Hi, we are developing and testing our new "NVMe SSD" and toolbox.
    When we try to get SMART info from our NVMe SSD at Windows 8.1, we are having troubles that Windows 8.1 seems not pass "Get Log page" command which is one of NVMe admin commands.
    However, we can successfully get SMART info from SATA-based SSD without any issues.
    So, just wondering if Windows 8.1 natively support all NVMe admin commands, or we need to develop our own special driver instead of Windows in-box driver (StorNVMe.sys).
    Thanks in advance for any help.
    BR, 
    HJ

    Hi HJ,
    StorNVMe.sys supports the admin commands internally. On win8.1, the miniport driver doesn't allow NVMe pass-though commands thus you cannot get NVMe log out of the driver.
    Regards,
    Michael Xing [MSFT]
    Thanks, Michael!
    Is there any other walk around to get NVMe log out of the driver?
    I found an old thread (below) that you mentioned SCSI pass through IOCTL commands, could you give me some more detailed information on how to do it?
    https://social.msdn.microsoft.com/Forums/windowsdesktop/en-US/8882bd78-b12e-43e3-a892-92a044ce4882/how-should-we-manage-the-nvme-devices-in-ws2012r2-?forum=wdk#f741642d-ec39-4fc2-a99c-e38bde20e7f9
    Thanks a lot!
    HJ

  • Weblogic.Admin command

    After I ran setEnv.sh, I used Java weblogic.Admin command for THREAD_DUMP, it produced error "Exception in thread "main" java.lang.NoClassDefFoundError: weblogic/Admin
    ". What would be the best way to run a THREAD_DUMP, kill -3 PID or a java command?

    hi sir,
    I think in weblogic 9.2 weblogic.Admin utility is deprecated but it works in 9.2. We have better option than weblogic.Admin utility nothing but WLST. If u want to start using weblogic.Admin we have command
    java weblogic.Admin -url adminhost:listenport -username xxxx -password xxxx START managedservername
    java weblogic.Admin -url adminhost:listenport -username xxxx -password xxxx SHUTDOWN managedservername
    java weblogic.Admin -url adminhost:listenport -username xxxx -password xxxx START clustername
    java weblogic.Admin -url adminhost:listenport -username xxxx -password xxxx SHUTDOWN clustername
    If u want to use WLST for starting managed servers and clusters
    check the follwing steps.
    1. java weblogic.WLST
    2. startNodeManager()
    3. TAKE ANOTHER COMMAND PROMPT START ADMIN SERVER
    4. TAKE ANOTHER COMMAND PROMPT
    java weblogic.WLST
    5. start('man1','Server') or start('man1','Server','localhost:7003')
    6. start('mycluster','Cluster')
    7.shutdown('man1','Server') or start('man1','Server','localhost:7003')
    8.shutdown('mycluster','Cluster')
    regards
    abhi
    Edited by: sumanth_abhi on Feb 5, 2009 8:54 PM

  • DBM_EXECUTE  command timeout exceeded

    Hello All,
    When i am excuting any db t-code from my system ,
    i am gettting DBM_EXECUTE  command timeout exceeded.
    I have tried it with different user but facing the same problem.
    kindly let me know the sloution for this.
    Thanks in advance,
    Atul

    Hello Atul,
    if you don't find a solution in the SAP notes then you should open a customer message.
    SAP support will then logon to your system and analyze the problem.
    Best regards,
    Melanie

  • HIS 2013 - Command Timeout - Dynamic DB2 Sendport

    Hi ,
        In one of our application (BizTalk 2013 R2), We are integrating with DB2 (Host Integration server 2013) using dynamic sendport. The default command timeout for DB2 is 30 seconds. This timeout can be increased in DB2 static sendports.
    But as we are using Dynamic sendport, we are not seeing such option in the context properties to set the command timeout.
    Any suggestions on how to set command timeout for dynamic DB2 sendport?
    Thanks,
    Karthik M
    karthik

     This timeout can be increased in DB2 static sendports. But as we are using Dynamic sendport, we are not seeing such option in the context properties to set the command timeout.
    As you said, we really can't set timeout for Dynamic send port, you can consider
    contacting Microsoft support directly(http://support.microsoft.com)

  • SQL error - 20013, Admin command cannot enter critical state 'Backup'

    Hello,
    I have a problem to start (ADMIN --> ONLINE) our MaxDB of our Solution Manager.
    I get the following error message:
    SQL error
    -104, DBM command impossible at this time
    - 20013, Admin command cannot enter critical state 'Backup'
    Has someone an idea how to solve this issue?
    Thanks in advance for help!
    Kind regards,
    Joern

    Hi,
    that's my problem. I can't start the database:
    af4lm139:sm2adm 51> dbmcli -d SM2 -u control,xxx db_admin
    OK
    af4lm139:sm2adm 51> dbmcli -d SM2 -u control,xxx db_warm
    Can not flush file '/sapdb/SM2/data/wrk/dbmsrv_af4lm139.err'.
    dbmsrv_af4lm139.err:
    PID 598:      -
    PID 598:  25: RTEThread_Thread::AppointMainThreadToThreadObject(int, SAPDBErr_MessageList&) + 0x392
    PID 598:          Symbol: ZN16RTEThreadThread31AppointMainThreadToThreadObjectEiR20SAPDBErr_MessageList
    PID 598:          SFrame: IP: 0x000000000087d8f2 (0x000000000087d560+0x392)
    PID 598:          Module: /sapdb/SM2/db/pgm/dbmsrv
    PID 598:      -
    PID 598:  26: main + 0x105
    PID 598:          SFrame: IP: 0x000000000040bf15 (0x000000000040be10+0x105)
    PID 598:          Module: /sapdb/SM2/db/pgm/dbmsrv
    PID 598:      -
    PID 598:  27: __libc_start_main + 0xf4
    PID 598:          SFrame: IP: 0x00002adc11d1a304 (0x00002adc11d1a210+0xf4)
    PID 598:          Module: /lib64/libc-2.4.so
    PID 598:      -
    PID 598:  28: __gxx_personality_v0@@CXXABI_1.3 + 0x10a
    PID 598:          SFrame: IP: 0x000000000040bd5a (0x000000000040bc50+0x10a)
    PID 598:          Source: start.S:116
    PID 598:          Module: /sapdb/SM2/db/pgm/dbmsrv
    Kind regards,
    Joern

  • Sudo chown root:admin / command information

    I would like to know if the use of this command before an OSX update would affect the operation of my computer after the 10.6.7 update. I had to use this command (suggested by vea1083) to fix a directory ownership disassociation following the iTunes 10.2.1 Update which slowed my MacBook Pro boot up speed by twice of the time. The following command was used to fix the issue and it worked successfully:
    *sudo chown root:admin /*
    Thank you for any assistance in this matter.

    Thank You for your responses, besides noticing that my thread has been moved to another forum let me give you a little insight on my decision to call in this command in the terminal window.
    It all started 1 month ago, I performed an iTunes update and to my surprise, my boot time went up from 22 seconds to almost 50 seconds. I cleared my log in items and repaired my permissions, with no success, so I had to take the sledgehammer approach and reinstall OSX and get the combo update to bring back my computer to OSX 10.6.6, after performing the reinstallation, my boot time went back to normal (22 seconds). Thinking that this was only a one time issue, when iTunes 10.2.0 came out I performed the iTunes update and again I was presented with the slow boot up issue after updating iTunes, angered by the fact that the same issue resurfaced, again I performed another permissions repair and reboot, and using my Mac in Safeboot mode failed, I had to again reinstall OSX again. Now hopeless, when the iTunes update 10.2.1 update came out, I did again the freaking update and boom again the same problem only this time I refused to reinstall OSX (I deny to see myself reinstalling OSX every single time I had to update iTunes). Fortunately I found a fix candidate in this Macrumors thread (http://forums.macrumors.com/showthread.php?t=1109887) where some MacBook Airs were reporting a similar boot time slowdown after the users performed an iTunes Update. Here was there where one user got the suggested the *sudo chown root:admin /* command:
    http://rifers.org/blogs/gbevin/2009/9/20/slowstartupshutdownsnowleopard
    I had hold off my 10.6.7 update pending a before update permissions fix and time machine backup to keep my computer from running into more troubles that I am already had.
    Thank you for your understanding and suggestions

  • Qlogic scsi command timeout

    Hi,
    I am doing some testing for Windows 2008 R2 SP1 boot from SAN with Qlogic HBAs.
    We noticed that a logical path failure will hang IO on the server for about 30 seconds...
    From traces it looks like 20 seconds is some king of timeout waiting to be expired (I suspect some kind of scsi command timeout)
    Another 7 seconds is spent with MPIO to disable that logical path...
    I tried searching the internet for such a timeout on windows and could not find any.
    Any of you know about this? Is there a way to verify this parameter value?
    Thanks
    Offer Baruch

    Hi,
    I am doing some testing for Windows 2008 R2 SP1 boot from SAN with Qlogic HBAs.
    We noticed that a logical path failure will hang IO on the server for about 30 seconds...
    From traces it looks like 20 seconds is some king of timeout waiting to be expired (I suspect some kind of scsi command timeout)
    Another 7 seconds is spent with MPIO to disable that logical path...
    I tried searching the internet for such a timeout on windows and could not find any.
    Any of you know about this? Is there a way to verify this parameter value?
    Thanks
    Offer Baruch
    You may increase default 30 seconds timeout to whatever you want (don't go too high however, 180 sounds like a maximum working one). 
    Changing the disk timeout on Microsoft Windows Server
    Learn how to change the disk timeout value on Windows Server 2000, 2003, and 2008 operating systems.
    On your Windows Server hosts, change the disk I/O timeout value to
    xxx in the Windows registry, as follows:
    In Windows, click Start > Run.
    In the dialog text box, type <samp class="ph codeph">regedit</samp> and click Enter.
    In the registry browsing tool, locate the HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Disk\TimeOutValue key.
    Confirm that the value for the key is xxx (decimal value). If necessary, change the value to
    xxx.
    You can read more info here:
    http://blogs.msdn.com/b/san/archive/2011/09/01/the-windows-disk-timeout-value-understanding-why-this-should-be-set-to-a-small-value.aspx
    Hope this helped. Good luck!
    StarWind VSAN [Virtual SAN] clusters
    Hyper-V without SAS, Fibre Channel, SMB 3.0 or iSCSI, uses
    Ethernet to mirror internally mounted SATA disks between hosts, starts with just a pair of boxes and scales-out to infinity.

  • FailedSend Playback Command Timeout!

    hi all
    iam getting the below error after clicking on the distributions button in the invoice workbench form when i am executing from command line.
    Failed     Send Playback Command Timeout! command:forms.button("//forms:button[(@name='LINE_SUM_CONTROL_DISTRIBUTIONS_0')]").click(), timeout: 10000
    can anybody tell me the fix for this issue it is little bit urgent

    1. add Thread.sleep(10000); before the click
    2. you can also try adding
    forms.button("//forms:button[(@name='LINE_SUM_CONTROL_DISTRIBUTIONS_0')]").waitFor(5000);
    before
    forms.button("//forms:button[(@name='LINE_SUM_CONTROL_DISTRIBUTIONS_0')]").click();
    3. verify the attributes and change "or" between attributes to "and"
    in any case if its not working, plz add 2-3 lines of code before and after the problematic line,and attributes of the button. it would help understand issue better.

  • Setting timeout in weblogic.admin command

    Hello:
    I run a script that checks the state of several thing in my platform. One of the commands I use in this script
    is
    java -cp .. weblogic.Admin localhost:7001 ..... GETSTATE
    IS there anyway to set a timeout in this command for the case weblogic server is hang and doesn´t answer anything. I have tried using sun properties for connect timeout and readTimeout but they haven´t worked(I suppose that it is because weblogic.admin uses t3 instead http). I have thought to make the following
    sleep 10|ps -efx| grep -i GETSTATE|grep -v grep|awk '{ print $2 }'|xargs kill -9 &
    java -cp .. weblogic.Admin localhost:7001 ..... GETSTATE
    but I don´t like at all this solution
    Regards,
    Mario.

    hi sir,
    I think in weblogic 9.2 weblogic.Admin utility is deprecated but it works in 9.2. We have better option than weblogic.Admin utility nothing but WLST. If u want to start using weblogic.Admin we have command
    java weblogic.Admin -url adminhost:listenport -username xxxx -password xxxx START managedservername
    java weblogic.Admin -url adminhost:listenport -username xxxx -password xxxx SHUTDOWN managedservername
    java weblogic.Admin -url adminhost:listenport -username xxxx -password xxxx START clustername
    java weblogic.Admin -url adminhost:listenport -username xxxx -password xxxx SHUTDOWN clustername
    If u want to use WLST for starting managed servers and clusters
    check the follwing steps.
    1. java weblogic.WLST
    2. startNodeManager()
    3. TAKE ANOTHER COMMAND PROMPT START ADMIN SERVER
    4. TAKE ANOTHER COMMAND PROMPT
    java weblogic.WLST
    5. start('man1','Server') or start('man1','Server','localhost:7003')
    6. start('mycluster','Cluster')
    7.shutdown('man1','Server') or start('man1','Server','localhost:7003')
    8.shutdown('mycluster','Cluster')
    regards
    abhi
    Edited by: sumanth_abhi on Feb 5, 2009 8:54 PM

Maybe you are looking for

  • ITunes wont open, getting error message, hard drive crash

    i had my secondary hard drive die. My windows is on my C: all of my music was on my D:. Now iTunes wont open becuase it cant access D:. If I try to uninstall it tells me it cant find D:. If I try to re-install it says it cant find D:. I have cleaned

  • Midori segmentation fault

    When browsing with midori, it sometimes crashes. It seems to be random, a page might chras it someday, but not the otherday. When run in terminall with -g it gives this: Launching command: '/usr/bin/gdb' --batch -ex 'set print thread-events off' -ex

  • Upgrade from 8.1 SP 6 to 9.2 or 10.0?

    We use Weblogic v8.1 SP 6 in production and want to upgrade. Is it too early for v10.0 and should we go with v9.2? Wait for v10.1 (how long will that take)? Greetings, Peter

  • How to get current resolution of the window

    I want set my JFrame's size according to the window resolution, but how can I get it? Thanks!

  • CS4 Define Pattern Acting Weirdly!!! Pleas Help (with screenshot)!!!

    I'm try to define a 20x20 pixel pattern for infill to create an old crt monitor effect. But Photoshop won't let me define the whole 20x20 but only allows me a 20x1 strip. (I've selected the whole image) There's also a weird "16" at the corner of the