function readOnly(count){ }
Starting November 20, the site will be set to read-only. On December 4, 2023,
forum discussions will move to the Trailblazer Community.
+ Start a Discussion
QueballQueball 

Connection Reset

Hello All,

Recently I have been noticing a Connection Reset error being thrown from the SFDC API.  Durring night-time hours I have noticed that this error doesn't occur as often.  This leads me to belive that the SFDC API is under considerable load durring daytime hours and that this is effecting the rate of return of records from SFDC.  Is this a correct assumption?  If you need a more specific description of our problem please ask.

Regards,
Jason D. Peterson

Message Edited by Queball on 02-14-2007 03:15 PM

Rick.BanisterRick.Banister
You are correct in your assumption. While the API is well behaved in optimal conditions, the devil's in the details. It takes significant effort and lots of heavy usage experience to find and handle every possible error condition in a complex API application, including data integrity issues, SOAP stack failures, network disconnects, and planned or unplanned Salesforce.com outages. Our integration tool, Relational Junction, reports on connect failures, attempts reconnection for up to 20 minutes, and restarts where it left off. Historically, the big improvement in Salesforce.com reliability came about in June 2006 when Salesforce.com split its server farm into multiple servers.

Realistically, ANY computer-based technology can have some degree of failure and most mature applications are 80-90% error checking logic. I'm forever glad that Salesforce.com takes care of its operations as well as they do, because the reliability is better than any self-hosted system I've ever seen.
benjasikbenjasik
would be great if you could log a ticket with the details of what's happening so we can help you troubleshoot it
QueballQueball
Thanks for the help.

The curious thing is that sometime late last week most of the connection issues no longer showed their ugly head. :smileyhappy:  While I do anticipate outages and "slowdowns" (we are talking a growing on demand platform of course), this issue lasted almost a week in duration.

I will post a support request anyway.  We would like to know what the problem was and how we can better work with SFDC in future circumstances