Research Academy
My account Customers About Blog Careers Legal Contact Resellers

Burp Suite User Forum

Create new post

Enterprise Server Crashing Once or Twice a Day

Nathan | Last updated: Oct 18, 2023 08:15AM UTC

Support, We have been experiencing issues with Burpsuite Enterprise since early/mid September where the Web console becomes inaccessible and access attempts timeout. We have the 'server'/'manager' components installed on a single m5a.xlarge EC2 instance with distributed scanners. As we monitor our Enterprise deployment, we've seen a noticeable uptick in regular CPU spikes (approx every 30-mins) since mid September. There has been no significant change in scanning activities over the last 6 months and prior to September the CPU activity was consistently 'stable' (very little spiking) - we can provide CPU charts to demonstrate this. During initial investigation it was seen that a java processes was maxing out the CPUs every 30 mins for a approx 5 mins. After a few cycles of 5minute-long max CPU spikes, the CPU would then remain at maximum and access to the Manager WebUI console would fail - we can provide CPU charts to demonstrate this also. To remedy this failure each time, we would have to restart the Manager's Enterprise and Web services. Once done the server returns to normal behaviour and the WebUI console is accessible. To troubleshoot further, we migrated what was an embedded database to MySQL yesterday (17th Oct) thinking this was an issue (our embedded DB had grown quite large and your recommendation was to move to an external DB in production so we successfully moved to MySQL). Prior to the DB migration, we did attempt to uninstall and reinstall Enterprise but this made no difference to the behaviour. Overnight, after the DB migration, the same failure occurred and the CPU usage 'pattern' suggests the same behaviour - several cycles of 5min-long max CPU spikes then a failure; with services needing to restart to fix. A number of java-related errors appear in the webserver.log and enterpriseServer.log Here's our setup: - A single m5a.xlarge EC2 instances runs the Manager components - DB, WebServer and Enterprise services. - Manager is Enterprisev2023.9.1 on Ubuntu 20.04.6 LTS - Manager Java version is openjdk 17.0.8 as taken from the /jre/bin directory - 6 x c6i.xlarge EC2 instances run the 6 x licensed scanners running v2023.10.1.1 on Ubuntu 20.04.6 LTS Can you provide any assistance with this problem please? I am happy to provide an logs and monitoring info we have gathered. Thanks Nathan

Alex, PortSwigger Agent | Last updated: Oct 18, 2023 08:37AM UTC

Hi Nathan, Thanks for your post - I'm sorry to hear you are having stability issues with your deployment. To investigate this issue further for you, can you provide the support pack logs for review: Support pack logs: Help Center (?) > Support Pack > Download. Additionally, the CPU charts you have referenced would be great as well. Am I correct in saying this issue only applies to the Enterprise/web server - the scanning machines do not observe the same behavior? And do the spikes in resource consumption coincide with running scans? You can submit everything to support@portswigger.net and mark it for my attention. Once I've reviewed everything, I should be able to advise further. Best regards,

You must be an existing, logged-in customer to reply to a thread. Please email us for additional support.