I have a Django app running on Apache with wsgi module.
After few hours of inactivity I get that error and I have to restart the Apache.
Any ideas?
Thanks
This error message mean that the database server has closed the connection to you. I guess this is caused because the connection is idle.
I believe you can get this fixed by adjusting the wait_timeout inside the configuration file of your mysql database server. The file is most commonly named "my.cnf".
This, however, is not considered as a good practice. I would like to suggest you to optimize the application you are writing to open the connection to mysql on demand - there's no point to keep it open if you are not actively using it for a long time.
If you need a quick fix, use the mysql_ping() function to check whether the connection is still alive and re-open if necessary.
I'm running a MsAccess with Mysql background database and today I am experiencing some problems.
Depending on how far I can get into the program I am able to get a list of all the names and entry. But once I go into the edit form of any entry I get a variety of results. Sometimes I am lucky and I can see the first entry. But any entry after that will always get a MsAccess error: Object invalid or no longer set and then any subsequent calls after will yield a Mysql ODBC error of Mysql server has gone away.
I've looked around at several websites and even all the stack overflow sites with the same question and I've tried a variety of solutions. (Keep in mind that this database has been running for years and this is the first time I am getting this message) It also takes about 2m14s to 2m59s before it displays any error messages.
Here is what I tried so far:
I've tried changing a few of the ODBC connection settings.
Checked Allow Big Results - FAILED
Checked Enable Auto Reconnect - FAILED
I've Checked and repaired tables - FAILED
A good number of tables resulted in, "table needs update please repair table"
I've Doubled the numbers in the my.cnf configuration and restarted the mysql service as well as the msaccess software. FAILED
These were the settings of my my.cnf after I made the changes.
Here:
set-variable = max_connections=500
safe-show-database
log-error=/var/log/mysqld.log
connect_timeout=1000
interactive_timeout=28800
wait_timeout=288000 <**This was changed**
join_buffer_size=6M
key_buffer_size=300M <**THIS WAS CHANGED **
max_allowed_packet=300M <**THIS WAS CHANGED **
myisam_sort_buffer_size=300M <**THIS WAS CHANGED **
read_buffer_size=6M
sort_buffer_size=6M
table_cache=12288
thread_cache_size=24
tmp_table_size=132M
query_cache_limit=3M
query_cache_size=64M
query_cache_type=1
Basically I've tried every suggestion I could find so far and I can't seem to figure out the problem.
I've also had a look at MYSQL: Has gone away
If I go through the bulleted list there
Nobody has killed the running thread
I don't think the query was ran after the connection was closed (since it ran for years fine)
Client application does have the privileges needed
I don't know how to figure out if I had a timeout from the TCP/IP connection on client side
I don't know if I've encounted a timeout on the server side, but I do know that automatic reconnection in client is disabled
???
The query could potentially be large since it is a form with many subforms
???
DNS should be ok, since I can connect to it with a real sql viewer (HeidiSQL)
???Child forks???
???
I think this is one of those bugs that is hard to figure out since I'm exhausted :/ I'm probably missing some info but I am not sure what else to include.
---EDIT---
Thank you all for your comments, I'm still debugging this issue. It seems that it's not all the forms that are causing some issues. So I'm starting to think that this is a MSAccess issue more than it is an MySql issue. The forms that do break all have the same line in their VBA code:
Private Sub Form_Error(DataErr As Integer, Response As Integer)
If Screen.ActiveControl.Name = "UnboundTextBox" Then
Response = acDataErrContinue
End If
End Sub
I don't recall putting this code in, so I assume it's an automatic thing, but I'm still going through all my forms to see which ones cause this error and which do not. So more in a few.
---Today I'm doing some debugging trying to find whatever I can from whatever log I can muster ----
MSACCESS 17ec-b10 EXIT SQLStatisticsW with return code -1 (SQL_ERROR)
HSTMT 0AF82920
WCHAR * 0x00000000 [ -3] <empty string>
SWORD -3
WCHAR * 0x00000000 [ -3] <empty string>
SWORD -3
WCHAR * 0x0013AAE8 [ -3] "location\ 0"
SWORD -3
UWORD 1 <SQL_INDEX_ALL>
UWORD 0 <SQL_QUICK>
DIAG [08S01] [MySQL][ODBC 3.51 Driver][mysqld-5.0.92-community]Lost
connection to MySQL server during query (2013)
----------------------------- FINAL EDIT --------------------------
The past week or so, my boss has been working hard to install a new internal server for our office. We also switched IPs several times and ISP providers. Turns out, as a result he created for me an administrative nightmare, as I attempted to debug issues with the server, and issues with the software ect.. ect.. when the issue was in the settings the ISP providers established with the router settings etc... etc... SO basically this whole headache was nothing wrong with server, nor software but hardware inbetween.
So now, I have to backwards engineer everything I did and try to get it to work on both ips. It works on the important one now. But it's not working on the old one so... I guess I just have to figure that out. [Waving imaginary fist in the air] Lol
The problem is that the connection is timing out. This is not a setting that you can change at the ODBC level. My MySQL provider had set the timeout at 30 seconds (you can find out what yours is set at my running the PassThrough query "SELECT ##wait_timeout"). I got round it by polling the connection every 10 seconds with a simple form. That keeps the connection alive. Make sure you have an AutoExec macro which opens the form.
Private Sub Form_Open(Cancel As Integer)
Me.TimerInterval = 1000 ' 10 seconds
End Sub
Private Sub Form_Timer()
Me.Requery
End Sub
I have not used MySQL with Access in a production environment, but the problem you are having sounds suspiciously similar to one I've encountered with SQL Server. Basically, if the client workstation has incorrect DNS settings, it can get confused about how to resolve the connection to the SQL Server and lose the connection. In the cases where I've encountered this, the workstations were incorrectly configured with the Internet provider's DNS as primary DSN, but the SQL Server was a local server, available only on the local LAN (and not mapped by the ISPs DNS).
The solution is to use the local domain controller as the primary DNS. It will look up the local SQL Server's IP and pass off any non-local lookups to the Internet provider's DNS.
If you don't have a Windows/Samba domain controller, then you need to use whatever the local LAN's authorative name server is. Alternatively, you could connect to the MySQL server by IP address and avoid the problem, or map it in the client workstation's HOSTS file (which is not recommended -- it's a great way to cause really hard to troubleshoot problems when the IP addresses change).
[mysqld-5.0.92-community]Lost
connection to MySQL server during query (2013)
This genary means you have found a bug in MySQL-Server
can you post a tail of your mysqld.err and mysqd.log
this makes the server restart and will effectefly shut down all open connections
Occasionally we're seeing an error from ASP pages:
Microsoft OLE DB Provider for ODBC Drivers error '80004005'
[MySQL][MyODBC 5.00.11][MySQL] 2006 MySQL server has gone away
We're handling it the best way we can in ASP but it still crops up. I think it's more to do with the MySQL ODBC driver than the ASP code. We never see this with PHP code we also have running under the same IIS server, however PHP re-connects each time it is ran, whereas I believe the ODBC driver connects once and stays connected.
I've checked the settings in the driver but there doesn't seem to be anything I can change to help mitigate the problem.
Can anyone explain why this is happening and how to reduce the number of times it happens?
The problem is that the connection is timing out. This is not a setting that you can change at the ODBC level. I got round it by polling the connection every 10 seconds with a simple query. That keeps the connection alive.
Not sure about ASP, but in Java/Tomcat/DBCP we have ability to stick a simple test statement (such as SELECT 1) in connection pools before actually getting the connection. Maybe there is something similar in ASP, too?
Closed. This question is not reproducible or was caused by typos. It is not currently accepting answers.
This question was caused by a typo or a problem that can no longer be reproduced. While similar questions may be on-topic here, this one was resolved in a way less likely to help future readers.
Closed 6 years ago.
Improve this question
I'm getting a SQL Server error:
A transport-level error has occurred
when receiving results from the
server. (provider: Shared Memory
Provider, error: 0 - The handle is
invalid.)
I'm running Sql Server 2008 SP1, Windows 2008 Standard 64 bit.
It's a .Net 4.0 web application. It happens when a request is made to the server. It's intermittent. Any idea how I can resolve it?
The database connection is closed by the database server. The connection remains valid in the connection pool of your app; as a result, when you pickup the shared connection string and try to execute it's not able to reach the database. If you are developing Visual Studio, simply close the temporary web server on your task bar.
If it happens in production, resetting your application pool for your web site should recycle the connection pool.
Try the following command on the command prompt:
netsh interface tcp set global autotuning=disabled
This turns off the auto scaling abilities of the network stack
I had the same problem. I restarted Visual Studio and that fixed the problem
Transport level errors are often linked to the connection to sql server being broken ... usually network.
Timeout Expired is usually thrown when a sql query takes too long to run.
So few options can be :
Check for the connection in VPN (if used) or any other tool
Restart IIS
Restart machine
Optimize sql queries.
For those not using IIS, I had this issue when debugging with Visual Studio 2010. I ended all of the debugger processes: WebDev.WebServer40.EXE which solved the issue.
All you need is to Stop the ASP.NET Development Server and run the project again
If you are connected to your database via Microsoft SQL Server Management, close all your connections and retry.
Had this error when connected to another Azure Database, and worked for me when closed it.
Still don't know why ..
Look at the MSDN blog which details out this error:
Removing Connections
The connection pooler removes a connection from the pool after it has
been idle for a long time, or if the pooler detects that the
connection with the server has been severed.
Note that a severed connection can be detected only after attempting
to communicate with the server. If a connection is found that is no
longer connected to the server, it is marked as invalid.
Invalid connections are removed from the connection pool only when
they are closed or reclaimed.
If a connection exists to a server that has disappeared, this
connection can be drawn from the pool even if the connection pooler
has not detected the severed connection and marked it as invalid.
This is the case because the overhead of checking that the connection
is still valid would eliminate the benefits of having a pooler by
causing another round trip to the server to occur.
When this occurs, the first attempt to use the connection will detect
that the connection has been severed, and an exception is thrown.
Basically what you are seeing is that exception in the last sentence.
A connection is taken from the connection pool, the application does
not know that the physical connection is gone, an attempt to use it is
done under the assumption that the physical connection is still there.
And you get your exception.
There are a few common reasons for this.
The server has been restarted, this will close the existing connections.
In this case, have a look at the SQL Server log, usually found at:
C:\Program Files\Microsoft SQL Server\\MSSQL\LOG
If the timestamp for startup is very recent, then we can suspect that
this is what caused the error. Try to correlate this timestamp with
the time of exception.
2009-04-16 11:32:15.62 Server Logging SQL Server messages in file
‘C:\Program Files\Microsoft SQL Server\MSSQL.1\MSSQL\LOG\ERRORLOG’.
Someone or something has killed the SPID that is being used.
Again, take a look in the SQL Server log. If you find a kill, try to
correlate this timestamp with the time of exception.
2009-04-16 11:34:09.57 spidXX Process ID XX was killed by
hostname xxxxx, host process ID XXXX.
There is a failover (in a mirror setup for example) again, take a look in the SQL Server log.
If there is a failover, try to correlate this timestamp with the time
of exception.
2009-04-16 11:35:12.93 spidXX The mirrored database “” is changing roles from “PRINCIPAL” to “MIRROR” due to
Failover.
Was getting this, always after about 5 minutes of operation. Investigated and found that a warning from e1iexpress always occurred before the failure. This apparently is an error having to do with certain TCP/IP adapters. But changing from WiFi to hardwired didn't affect it.
So tried Plan B and restarted Visual Studio. Then it worked fine.
On closer study I noticed that, when working correctly, the message The Thread '<No Name>' has exited with code 0 occurred at almost exactly the time the run crashed in previous attempts. Some Googling reveals that that message comes up when (among other things) the server is trimming the thread pool.
Presumably there was a bogus thread in the thread pool and every time the server attempted to "trim" it it took the app down.
You get this message when your script make SQL Service stopped for some reasons. so if you start SQL Service again perhaps your problem will be resolved.
I know this may not help everyone (who knows, maybe yes), but I had the same problem and after some time, we realized that the cause was something out of the code itself.
The computer trying to reach the server, was in another network, the connection could be established but then dropped.
The way we used to fix it, was to add a static route to the computer, allowing direct access to the server without passing thru the firewall.
route add –p YourServerNetwork mask NetworkMask Router
Sample:
route add –p 172.16.12.0 mask 255.255.255.0 192.168.11.2
I hope it helps someone, it's better to have this, at least as a clue, so if you face it, you know how to solve it.
I got the same error in Visual Studion 2012 development environment, stopped the IIS Express and rerun the application, it started working.
I had the same issue. I solved it, truncating the SQL Server LOG.
Check doing that, and then tell us, if this solution helped you.
For me the solution was totally different.
In my case I had an objectsource which required a datetimestamp parameter. Even though that ODS parameter ConvertEmptyStringToNull was true 1/1/0001 was being passed to SelectMethod. That in turn caused a sql datetime overflow exception when that datetime was passed to the sql server.
Added an additional check for datetime.year != 0001 and that solved it for me.
Weird that it would throw a transport level error and not a datetime overflow error.
Anyways..
In my case the "SQL Server" Server service stopped. When I restarted the service that enabled me to run the query and eliminate the error.
Its also a good idea to examine your query to find out why the query made this service stop
For me the answer is to upgrade the OS from 2008R2 to 2012R2, the solution of iisreset or restart apppool didn't work for me.
I also tried to turn of TCP Chimney Offload setting, but I didn't restart the server because it is a production server, which didn't work either.
We encountered this error recently between our business server and our database server.
The solution for us was to disable "IP Offloading" on the network interfaces.
Then the error went away.
One of the reason I found for this error is 'Packet Size=xxxxx' in connection string. if the value of xxxx is too large, we will see this error. Either remove this value and let SQL server handle it or keep it low, depending on the network capabilities.
It happened to me when I was trying to restore a SQL database and checked following Check Box in Options tab,
As it's a stand alone database server just closing down SSMS and reopening it solved the issue for me.
This occurs when the database is dropped and re-created some shared resources is still considering the database still exists, so when you re-run execute query to create tables in the database after it was re-created the error will not show again and Command(s) completed successfully. message will show instead of the error message Msg 233, Level 20, State 0, Line 0 A transport-level error has occurred when sending the request to the server. (provider: Shared Memory Provider, error: 0 - No process is on the other end of the pipe.).
Simply ignore this error when you are dropping and recreating databases and re-execute your DDL queries with no worries.
I faced the same issue recently, but i was not able to get answer in google.
So thought of sharing it here, so that it can help someone in future.
Error:
While executing query the query will provide few output then it will throw below error.
"Transport level error has occurred when receiving output from
server(TCP:provider,error:0- specified network name is no longer
available"
Solution:
Check the provider of that linked server
In that provider properties ,Enable "Allow inprocess" option for that particular provider to fix the issue.