DepotConnectionFailureException: Error downloading bundle [/COMP/SDDC_MANAGER_VCF/bundles/bundle-<TBD>.tar] from dl.broadcom:443
search cancel

DepotConnectionFailureException: Error downloading bundle [/COMP/SDDC_MANAGER_VCF/bundles/bundle-<TBD>.tar] from dl.broadcom:443

book

Article ID: 402929

calendar_today

Updated On:

Products

VMware SDDC Manager

Issue/Introduction

  • SDDC Manager - Bundle Manager is failing to download bundles. Download task takes a long time to start.
  • Unable to download update through SDDC Manager.
  • The UI displays one of the following specific exceptions depending on the bundle and depot source:
  • DepotConnectionFailureException: Error downloading bundle [/COMP/SDDC_MANAGER_VCF/bundles/bundle-<TBD>.tar] from dl.broadcom:443
    DepotConnectionFailureException: Error downloading bundle [/evo/vmw/bundles/bundle-#####.tar] from depot.vmware.com:443


  • Reviewing the LCM debug log (/var/log/vmware/vcf/lcm/lcm-debug.log) reveals a Repo Write Failure
    DEBUG [vcf_lcm,ba0cd####e8#####1,ed##] [c.v.e.s.l.p.impl.vsan.VsanHealthUtil,http-nio-127.0.0.1-7##0-exec-6] Health query Result: HealthCheckQueryResult(resultId=0####45-d##c-##f3-b7##-83######2e49, status=SUCCESS, result=[])
    DEBUG [vcf_lcm,ba0cd####e8#####1,ed##] [c.v.v.l.r.a.c.v.v.HealthCheckController,http-nio-127.0.0.1-7400-exec-6] getHealthCheckStatus result: HealthCheckQueryResult(resultId=0####45-d##c-##f3-b7##-83######2e49, status=SUCCESS, result=[])
    ERROR [vcf_lcm,699323############7b7c49c1df,e8##f,bundleId=8b######3-d##d-###4-##9d-4#######9b6e] [c.v.e.s.l.f.b.c.i.BundleFileClientImpl,vac-scheduler-1] Failed to store bundle /nfs/vmware/vcf/nfs-mount/bundle/depot/local/bundles/bundle-251#####8.tar into the repository
    ERROR [vcf_lcm,699323############7b7c49c1df,e8af] [c.v.e.s.l.s.i.BundleDownloadExceptionHandlerImpl,vac-scheduler-1] Bundle download failed
    com.vmware.evo.sddc.lcm.model.bundle.repo.exception.BundleRepoWriteFailureException: Failed to store bundle(REPO WRITE FAILURE, path = /nfs/vmware/vcf/nfs-mount/bundle/depot/local/bundles/bundle-251#####8.tar
            at com.vmware.evo.sddc.lcm.file.bundle.client.impl.BundleFileClientImpl.storeBundle(BundleFileClientImpl.java:142)
    
  • The NFS mount path /nfs/vmware/vcf/nfs-mount shows high or near 100% disk utilization. We can verify the same using below command:
    df -h

 

Environment

VMware Cloud Foundation 5.2

Cause

Issue is the lack of space on /nfs/vmware/vcf/nfs-mount/ and when the SDDC Manager attempts to write the compressed bundle to the local repository, the underlying filesystem hits its capacity limit. This triggers a REPO WRITE FAILURE, as the BundleFileClientImpl cannot finalize the file transfer to the disk.

Behavior is reproducible via triggering a bundle via GUI and tail-ing the lcm-debug.log 

# tail -f /var/log/vmware/vcf/lcm/lcm-debug.log | grep -i "no space left on device"

Can also be listed via 
root@sddc_manager [ ~ ]# df -h
Filesystem                                   Size  Used Avail Use% Mounted on
devtmpfs                                     4.0M     0  4.0M   0% /dev
tmpfs                                        7.9G   32K  7.9G   1% /dev/shm
tmpfs                                        3.2G  8.8M  3.2G   1% /run
tmpfs                                        4.0M     0  4.0M   0% /sys/fs/cgroup
/dev/mapper/vg_system-lv_root                 26G   15G   10G  59% /
/dev/mapper/vg_lvm_snapshot-lv_lvm_snapshot   87G   24K   82G   1% /storage/lvm_snapshot
/dev/mapper/vg_alt_root-lv_alt_root           26G   24K   25G   1% /storage/alt_root
tmpfs                                        7.9G   40M  7.8G   1% /tmp
/dev/mapper/vg_lcmmount-lv_lcmmount          503G  227G  252G 100% /nfs/vmware/vcf/nfs-mount
/dev/mapper/vg_data-lv_data3                 165G  5.0G  151G   4% /opt/vmware/vcf
/dev/sda3                                    488M   48M  404M  11% /boot
/dev/sda2                                     10M  2.0M  8.1M  20% /boot/efi
/dev/mapper/vg_data-lv_data2                  63G   39G   21G  66% /var/log/vmware
/dev/mapper/vg_data-lv_data1                 7.8G  132M  7.3G   2% /data



Resolution

To resolve this issue, free up disk space by removing unused and old bundles from the NFS repository.

  1. SSH to the SDDC Manager VM with vcf user credentials and then switch to root login.
  2. Verify Disk Utilization using below command:
    df -h
  3. Run the following command to verify which specific subdirectory is consuming the most space:
    du -sh /nfs/vmware/vcf/nfs-mount/*
  4. List the contents of the bundle directory to find the IDs of older bundles that are no longer needed:
    ls -lh /nfs/vmware/vcf/nfs-mount/bundle/depot/local/bundles/

    Note: The Bundle ID associated with large files you wish to delete (e.g., if the file is bundle-12345.tar, the ID is 12345).

  5. List the files to identify older bundles that are no longer needed. Remove them using below command:
    /opt/vmware/vcf/lcm/lcm-app/bin/bundle_cleanup.py {bundle id}
    

    For multiple bundles, use the same CLI, as listed below

    python /opt/vmware/vcf/lcm/lcm-app/bin/bundle_cleanup.py "{Bundle id -#####}" "{Bundle id -#####}"

    Reference: How to clean up old/ installed bundles in SDDC manager using Bundle Cleanup Utility

  6. Run df -h again to confirm that sufficient space has been reclaimed. Log back into the SDDC Manager UI and retry the bundle download process.