mapr-support-dump.sh
This script collects node and cluster-level information about the
node where the script is invoked. The information collected is used
to help MapR Support diagnose problems. Use
mapr-support-collect.sh
to collect diagnostic information from all nodes in the
cluster.
The "mini-dump"
option limits the size of the support output. When the
-m
or --mini-dump
option is specified
along with a size, mapr-support-dump.sh
collects only
a head and tail, each limited to the specified size, from any log
file that is larger than twice the specified size. The total size
of the output is therefore limited to approximately 2 * size *
number of logs. The size can be specified in bytes, or using the
following suffixes:
- b - bytes
- k - kilobytes (1024 bytes)
- m - megabytes (1024 kilobytes)
Syntax
/opt/mapr/support/tools/mapr-support-dump.sh
[ -n| --name <name> ]
[ -l| --no-logs ]
[ --no-hadoop-logs ]
[ --hbase-logs ]
[ --sqoop-logs ]
[ --eco-logs ]
[ --oozie-logs ]
[ --spark-logs ]
[ --pig-logs ]
[ --impala-logs ]
[ --hue-logs ]
[ --hive-logs ]
[ --flume-logs ]
[ --drill-logs ]
[ -L| --libraries ]
[ -d| --output-dir <path> ]
[ -s| --no-statistics ]
[ -c| --no-conf ]
[ -i| --no-sysinfo ]
[ -o| --exclude-cluster ]
[ -O| --online ]
[ -z| --only-cluster ]
[ --collect-cores ]
[ --move-cores ]
[ --port <port> ]
[ --nfsport <port> ]
[ -m| --mini-dump <size> ]
[ -A| --logs-age <days> ]
[ -?| --help ]
Parameters
Parameter |
Description |
---|---|
-n or --name |
Specifies the name of the output file. If not specified, the default is a date-named file in the format YYYY-MM-DD-hh-mm-ss.tar |
-l or --no-logs | Does not include log files. |
--no-hadoop-logs | Does not include Hadoop log files |
--hbase-logs | Includes Hbase log files. |
--sqoop-logs | Includes Sqoop log files |
--eco-logs | Includes all ecosystem log files. |
--oozie-logs | Includes Oozie log files. |
--spark-logs | Includes Spark log files. |
--pig-logs | Includes Pig log files. |
--impala-logs | Includes Impala log files. |
--hue-logs | Includes Hue log files. |
--hive-logs | Includes Hive log files. |
--flume-logs | Includes Flume log files. |
--drill-logs | Includes Drill log files. |
-L or --libraries |
Includes libraries. |
-d or --output-dir |
The absolute path to the output directory. If not specified, the
default is |
-s or --no-statistics |
Does not include statistics. |
-c or --no-conf |
Does not include configurations. |
-i or --no-sysinfo |
Does not include system information. |
-o or --exclude-cluster |
Does not collect cluster diagnostics. |
-O or --online |
Saves the support dump output file to the /var/mapr/cluster/support directory in maprfs. When not specified, the output file is stored at /opt/mapr/support/dump on the local machine file system. Specifies a space-separated list of nodes from which to gather support output, and uses the warden instead of ssh for transmitting the support data. |
-z or --only-cluster | Collects diagnostic information at the cluster level only. |
--collect-cores |
Collects cores of running mfs processes from all nodes (off by default) |
--move-cores |
Moves mfs and nfs cores from /opt/cores from all nodes (off by default) |
--port |
The port number used by the FileServer. Default: 5660 |
--nfs-port |
The port used by the NFS server. Default: 9998 |
-m, --mini-dump <size> |
For any log file greater than 2 * <size>, collects only a head and tail each of the specified size. The <size> may have a suffix specifying units:
|
-A or --logs-age <days> |
Collects logs newer than the specified number of days. The
default value for this parameter is 7. Specify a value of 0 to have
the |
-? or --help |
Displays usage help text |
Output
The example produces a tar file at
/opt/mapr/support/dump/mysupport-output.tar
. To
extract the tar file, use tar -xf
mysupport-output.tar
. The directory structure is as
follows:
$ ls
mysupport-output mysupport-output.tar
$ ls mysupport-output
MapRBuildVersion hostid mfsstate_commands
cluster hostname roles
cluster_summary.txt linux-release support_dump.log
conf logs system_info
conf_file_metadata mapr-clusters.conf
-
mfsstate_commands - Contains all information collected about the state of the filesystem are stored the the mfsstate_commands directory.
- cluster - Contains information associated with jobtracker, maprcli_commands, resourcemanager, and volume_dumps.
- system_info - Contains all information related to the system state and configuration.
- support_dump.log - Contains console output of the individual commands launched by the support dump script.
- Commonly required files - cluster_summary.txt, conf_file_metadata, hostid, hostname, linux-release, MapRBuildVersion, mapr-clusters.conf, and support_dump.log are place in the dump root for easier access.
Example
Collect support information and dump it to the file /opt/mapr/support/collect/mysupport-output.tar:/opt/mapr/support/tools/mapr-support-dump.sh -n mysupport-output
Example Output
/opt/mapr/support/tools/mapr-support-dump.sh -n mysupport-output
2015-06-25 11:51:38.397 INFO Starting Support dump collection. For diagnostics, refer to supportèdump.log inside the dump
2015-06-25 11:51:38.402 INFO Collecting system information
2015-06-25 11:51:49.596 INFO Collecting mapr logs
2015-06-25 11:52:00.734 INFO Log collection from maprfs is succesful.
2015-06-25 11:52:00.804 INFO Collecting cluster configuration
2015-06-25 11:53:07.626 INFO Skipping /opt/mapr/logs/prerequisitecheck.log since it does not exist
2015-06-25 11:53:07.639 INFO Skipping /opt/mapr/hbase/hbase-0.94.17/conf since it does not exist
2015-06-25 11:53:10.436 INFO Collecting cluster summary information
2015-06-25 11:53:19.183 INFO Collecting detailed cluster information
2015-06-25 11:54:03.219 INFO Dump collection is succesful. Tar file: /opt/mapr/support/dump/mysupport-output.tar