Bug 1898097
| Summary: | mDNS floods the baremetal network | ||||||
|---|---|---|---|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Yuval Kashtan <ykashtan> | ||||
| Component: | Networking | Assignee: | Ben Nemec <bnemec> | ||||
| Networking sub component: | mDNS | QA Contact: | Oleg Sher <osher> | ||||
| Status: | CLOSED ERRATA | Docs Contact: | |||||
| Severity: | high | ||||||
| Priority: | high | CC: | afasano, asegurap, augol, bbennett, dblack, jlema, kseremet, smalleni | ||||
| Version: | 4.6 | Keywords: | Triaged | ||||
| Target Milestone: | --- | ||||||
| Target Release: | 4.7.0 | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: |
Cause: The library used to provide mDNS services in the cluster did not properly implement the mDNS protocol.
Consequence: Excessive multicast traffic was generated.
Fix: Limit multicast frequency to once per second.
Result: Significantly reduced multicast traffic.
|
Story Points: | --- | ||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2021-02-24 15:33:27 UTC | Type: | Bug | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Bug Depends On: | |||||||
| Bug Blocks: | 1936539 | ||||||
| Attachments: |
|
||||||
*** Bug 1898101 has been marked as a duplicate of this bug. *** The reason I opened 2 distinct BZs is that for the flood bug, even changing the frequency would be a solution but for the scalability one, really need a different solution (than mDNS, or at least it's current implementation) also, the other BZ include some (very) good discussion which we're missing here. Even the temporary fix mentioned in https://bugzilla.redhat.com/show_bug.cgi?id=1898101#c4 does not fix the issue. Please see: https://bugzilla.redhat.com/show_bug.cgi?id=1898101#c19 Also can we use https://bugzilla.redhat.com/show_bug.cgi?id=1898101 instead of this as the main tracking BZ as there is a lot of history in that. I had two nodes in the cluster that did not pick up the machineconfig change fue to bad nodeselectors and those were enough to DDoS the network :) . Fixing that, I see a drastic drop in the number of mcast packets. *** Bug 1893670 has been marked as a duplicate of this bug. *** Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: OpenShift Container Platform 4.7.0 security, bug fix, and enhancement update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2020:5633 |
Created attachment 1729743 [details] mdns tcpdump capture Version: 4.6 $ openshift-install version openshift-baremetal-install 4.6.0 Platform: IPI What happened? PnT Lab had to disconnect us from network due to excessive multicasts flooding the network. these multicasts are all mDNS traffic. see attachment for tcpdump What did you expect to happen? as all our servers have proper DNS records, we dont need mDNS. I'd want a install-config parameter to disable mDNS How to reproduce it (as minimally and precisely as possible)? just install a bunch of cluster on same broadcast domain and observe tcpdump