Setup a Cluster or HA proxy. What is your inform interval set to?
-dan > On Apr 27, 2018, at 1:06 PM, Sasan torabkheslat <[email protected]> wrote: > > Dear Zaid > > Thanks for your response, we have hundreds of devices configured to send > their data to GenieACS every minute. > > Is their any way to improve capacity of GenieACS to meet our needs? > > ------------------------------ > Sasan Torabkheslat > Linux Sysadmin > Mobile: +989034565053 > Personal Website: torabkheslat.ir Email: [email protected] > Professional Profile > > > > >> On Thu, Apr 26, 2018 at 10:06 PM, Zaid Abdulla <[email protected]> wrote: >> On Sat, 2018-04-21 at 16:41 +0430, Sasan torabkheslat wrote: >> > We saw Response 503 (Service Unavailable) error from GenieACS server >> > and the following message in debug mode: >> > >> > # RESPONSE Sat Apr 21 2018 16:38:49 GMT+0430 (+0430) >> > {"Retry-After":300} >> >> This means the server is overloaded. Your server might be getting a >> sudden surge of requests because a large number of devices are sending >> their periodic informs at the same moment. To solve this, randomize the >> value of PeriodicInformTime parameter to more evenly distribute the >> load. >> >> Something like this (not tested): >> >> const timeOffset = Math.trunc(Math.random() * 86400000); >> declare("InternetGatewayDevice.ManagementServer.PeriodicInformTime", >> null, {value: timeOffset}); >> >> -- >> Zaid Abdulla <[email protected]> >> _______________________________________________ >> Users mailing list >> [email protected] >> http://lists.genieacs.com/mailman/listinfo/users > > _______________________________________________ > Users mailing list > [email protected] > http://lists.genieacs.com/mailman/listinfo/users
_______________________________________________ Users mailing list [email protected] http://lists.genieacs.com/mailman/listinfo/users
