Apply Changes fails with "Review logs for failed jobs: clamd"

Products

• VMware Tanzu Gemfire

Issue/Introduction

Symptoms:
"Apply Changes" fails shortly after the Service VM for PCC is spun up and throws the following error.

Task 88492 | 21:20:05 | Updating instance server: server/68d991bd-2537-4054-8cf7-2f7c415f73c6 (0) (canary) (00:11:57)
                      L Error: 'server/68d991bd-2537-4054-8cf7-2f7c415f73c6 (0)' is not running after update. Review logs for failed jobs: clamd
Task 88492 | 21:32:02 | Error: 'server/68d991bd-2537-4054-8cf7-2f7c415f73c6 (0)' is not running after update. Review logs for failed jobs: clamd

Task 88492 Started  Mon Jan 28 21:06:54 UTC 2019
Task 88492 Finished Mon Jan 28 21:32:02 UTC 2019
Task 88492 Duration 00:25:08
Task 88492 error

FreshClam.log (/var/vcap/sys/log) shows the following error:

ClamAV update process started at Wed Jan 30 20:28:43 2019
Reading CVD header (main.cld): WARNING: main.cld not found on remote server
Reading CVD header (main.cvd): OK
main.cvd is up to date (version: 58, sigs: 4566249, f-level: 60, builder: sigmgr)
Reading CVD header (daily.cld): WARNING: daily.cld not found on remote server
Reading CVD header (daily.cvd): OK
daily.cvd is up to date (version: 25343, sigs: 2228905, f-level: 63, builder: raynman)
Reading CVD header (bytecode.cld): WARNING: bytecode.cld not found on remote server
Reading CVD header (bytecode.cvd): OK
bytecode.cvd is up to date (version: 328, sigs: 94, f-level: 63, builder: neo)

Environment

Cause

The issue is caused by the default memory CGroup of ClamAV being too low to load the ClamAV virus definitions into memory and on-disk memory (Swap) being off in Pivotal Cloud Cache.

Resolution

This issue can be fixed by updating the ClamAV runtime config to include the following property: 

addons:
- name: clamav
  jobs:
  - name: clamav
    release: clamav
  properties:
    clamav:
      memory_limit: 1500000000

By default, ClamAV (until version 1.4.39), consumes 1GB of memory. Therefore, increasing the CGroup memory limit to 1.5GB (1500000000) will resolve the issue.