[MESOS-7748] Slow subscribers of streaming APIs can lead to Mesos OOMing.

• Type: Bug
• Status: Accepted
• Priority: Critical
• Resolution: Unresolved
• Labels: foundations mesosphere reliability
• Sprint: Mesosphere Sprint 58
• Story Points: 8
• Assignee: Unassigned
• Reporter: Alex R
• Shepherd: Benjamin Mahler
• Created: 30/Jun/17 13:21
• Updated: 09/Jan/19 18:57

Description

For each active subscriber, Mesos master / slave maintains an event queue, which grows over time if the subscriber does not read fast enough. As the number of such "slow" subscribers grows, so does Mesos master / slave memory consumption, which might lead to an OOM event.

Ideas to consider:

• Restrict the number of subscribers for the streaming APIs
• Check (ping) for inactive or "slow" subscribers
• Disconnect the subscriber when there are too many queued events in memory

Issue Links

• is blocked by: MESOS-7931 LibeventSSLSocket send does not expose progress.