| Subject | Re: FireBird Connect Error |
|---|---|
| Author | Adam |
| Post date | 2006-06-27T06:22:37Z |
--- In firebird-support@yahoogroups.com, Ghazanfar Ali
<ghazanfarali17@...> wrote:
error message "Unable to complete network request to
host "localhost". And when i check the firebird logs it contains:
look like this is a very common error.
A TCP/IP connection has no state (by definition). When you ping it,
all that says is that at the moment in time you issue the ping, it
receives a response from the server within an acceptable timeframe.
Each packet is routed separately, so it is possible that there are
intermittent 'connections' being dropped. The server will drop a
connection if it has not been able to send some data to it.
Underlying layers will drop the connection after an inactivity
timeout (usually 2hrs). Some firewalls etc also assume that no
traffic = connection dead, when in reality the over the wire protocol
in Firebird is just quiet when nothing is occurring.
Anyway, those 5 digit codes (eg 10053) are winsock error codes. You
can google to get the meaning of each.
I notice at 10:37 the guardian restarted the service. It would only
do this if the engine was not running, so you may need to look back a
bit more in the log to see the details of the crash. If the engine
restarts, all connections to that engine are lost (obviously).
Adam
<ghazanfarali17@...> wrote:
>some interval when i try to connect to the database it gives me an
> Hi Everyone,
>
> I am running Firebird 1.5.2 my machine and it works fine. But after
error message "Unable to complete network request to
host "localhost". And when i check the firebird logs it contains:
>\bin\fbserver.exe
> LK_NEW (Client) Sat Jun 17 10:37:01 2006
> Guardian starting: C:\Program Files\Firebird\Firebird_1_5
>but i just cannot seem to find the solution to this one and it doesnt
> LK_NEW (Client) Sat Jun 17 10:49:32 2006
> INET/inet_error: read errno = 10053
>
> LK_NEW (Client) Sat Jun 17 10:49:32 2006
> INET/inet_error: receive in try_connect errno = 203
>
> LK_NEW (Server) Sat Jun 17 10:49:32 2006
> INET/inet_error: accept errno = 10022
>
> LK_NEW (Server) Sat Jun 17 10:49:33 2006
> INET/inet_error: accept errno = 10022
>
> LK_NEW (Client) Sat Jun 17 10:49:33 2006
> INET/inet_error: read errno = 10053
>
> LK_NEW (Client) Sat Jun 17 10:49:33 2006
> INET/inet_error: receive in try_connect errno = 203
>
> I have searched the internet to find the solution ot this issue
look like this is a very common error.
>make sure the %windir%\system32\drivers\etc\hosts file is valid
> Things i have tried so far:
>
> 1. checked the service is running
> 2. checked the localhost is working (by pinging it) and checked to
A TCP/IP connection has no state (by definition). When you ping it,
all that says is that at the moment in time you issue the ping, it
receives a response from the server within an acceptable timeframe.
Each packet is routed separately, so it is possible that there are
intermittent 'connections' being dropped. The server will drop a
connection if it has not been able to send some data to it.
Underlying layers will drop the connection after an inactivity
timeout (usually 2hrs). Some firewalls etc also assume that no
traffic = connection dead, when in reality the over the wire protocol
in Firebird is just quiet when nothing is occurring.
Anyway, those 5 digit codes (eg 10053) are winsock error codes. You
can google to get the meaning of each.
I notice at 10:37 the guardian restarted the service. It would only
do this if the engine was not running, so you may need to look back a
bit more in the log to see the details of the crash. If the engine
restarts, all connections to that engine are lost (obviously).
Adam