- About this guide
- 1 Reporting software problems
- 2 Troubleshooting logging and debugging
- 2.1 Accessing configuration settings at runtime
- 2.2 Activating Ceph debugging at boot time
- 2.3 Accelerating log rotation
- 2.4 Monitoring memory utilization
- 2.5 Enable system, log, and debug settings
- 2.6 Logging kernel RBD and CephFS clients
- 2.7 Per-service and per-daemon events
- 2.8 Debugging a crashing process running in a container
- 3 Troubleshooting cephadm
- 3.1 Pausing or disabling cephadm
- 3.2 Checking cephadm logs
- 3.3 Accessing Ceph daemon logs
- 3.4 Collecting
systemdstatus - 3.5 Listing configured container images
- 3.6 Listing all downloaded container images
- 3.7 Running containers manually
- 3.8 Failing to infer CIDR network error
- 3.9 Accessing the admin socket
- 3.10 Deploying a Ceph Manager manually
- 3.11 Distributing a program temporary fix (PTF)
- 3.12 Failure When Adding Hosts with cephadm
- 3.13 Disabling automatic deployment of daemons
- 4 Troubleshooting OSDs
- 5 Troubleshooting placement groups (PGs)
- 5.1 Identifying troubled placement groups
- 5.2 Placement groups never get clean
- 5.3 Stuck placement groups
- 5.4 Peering failure of placement groups
- 5.5 Failing unfound objects
- 5.6 Identifying homeless placement groups
- 5.7 Only a few OSDs receive data
- 5.8 Unable to write data
- 5.9 Identifying inconsistent placement groups
- 5.10 Identifying inactive erasure coded PGs
- 6 Troubleshooting Ceph Monitors and Ceph Managers
- 7 Troubleshooting networking
- 8 Troubleshooting NFS Ganesha
- 9 Troubleshooting Ceph health status
- 10 Troubleshooting the Ceph Dashboard
- 11 Troubleshooting Object Gateway
- 12 Troubleshooting CephFS
- 13 Hints and tips
- 13.1 Identifying orphaned volumes
- 13.2 Adjusting scrubbing
- 13.3 Stopping OSDs without rebalancing
- 13.4 Checking for unbalanced data writing
- 13.5 Increasing file descriptors
- 13.6 Integration with virtualization software
- 13.7 Firewall settings for Ceph
- 13.8 Testing network performance
- 13.9 Locating physical disks using LED lights
- 13.10 Sending large objects with
radosfails with full OSD - 13.11 Managing the 'Too Many PGs per OSD' status message
- 13.12 Managing the 'nn pg stuck inactive' status message
- 13.13 Fixing clock skew warnings
- 13.14 Determining poor cluster performance caused by network problems
- 13.15 Managing
/varrunning out of space
- 14 Frequently asked questions
- A Mises à jour de la maintenance de Ceph basées sur les versions intermédiaires de « Pacific » en amont
- Glossaire
Copyright © 2020–2024 SUSE LLC et contributeurs. Tous droits réservés.
Sauf indication contraire, ce document est concédé sous licence Creative Commons Attribution-ShareAlike 4.0 International (CC-BY-SA 4.0) : https://creativecommons.org/licenses/by-sa/4.0/legalcode.
Pour les marques commerciales SUSE, consultez le site Web http://www.suse.com/company/legal/. Toutes les marques commerciales de fabricants tiers appartiennent à leur propriétaire respectif. Les symboles de marque commerciale (®, ™, etc.) désignent des marques de SUSE et de ses sociétés affiliées. Des astérisques (*) désignent des marques commerciales de fabricants tiers.
Toutes les informations de cet ouvrage ont été regroupées avec le plus grand soin. Cela ne garantit cependant pas sa complète exactitude. Ni SUSE LLC, ni les sociétés affiliées, ni les auteurs, ni les traducteurs ne peuvent être tenus responsables des erreurs possibles ou des conséquences qu'elles peuvent entraîner.