Downloads (PDF, HTML & XML formats)
[FishEye Knowledge Base]
A FishEye instance can be managed using the fisheyectl
script. Before running this script you need to ensure that you have set the JAVA_HOME environment variable, or that java
is on the path.
Unix usage:
/FISHEYE_HOME/bin/fisheyectl.sh command [options]
Windows usage:
\FISHEYE_HOME\bin\fisheyectl.bat command [options]
The command parameter can be one of run
, start
or stop
(see below). You can also find convenience scripts for running each of these commands, such as run.sh
or run.bat
.
The run
command starts FishEye. This command runs FishEye in the foreground. It does not fork a background process.
The start
command has the same options as run
, but starts FishEye in the background.
Windows: FishEye will be run in a separate cmd.exe
window.
Unix: FishEye will be run with nohup
and the console output will be redirected to $FISHEYE_INST/var/log/fisheye.out
.
Options (for both run and start):
| Load configuration from the file at |
| Do not print anything to the console. |
| Print extra information to the debug log. |
The following options are currently available, but will be removed at a later date:
| Specifies the number of spaces to use to represent a tab character. The default is 8. |
| When rendering the directory tree on some pages, FishEye calculates if each directory subtree is empty. For massive repositories, this calculation can cause the page to take a long time to render. This option disables the calculation that determines emptiness. If you start FishEye with this flag, then empty directories will no longer show up as grey (i.e. deleted) and the option to hide them will also disappear |
| Disable the generation of a full-text index for file content. This prevents further indexing, but does not delete any existing full-text indexes. FishEye will not warn you if you specify this option but still try to do a content search. This option is useful if you do not use content search and you are finding FishEye is taking a long time to index your content. |
--Xenable-git-content-hash-resolving-heuristic | When several git revisions have the same hash (for instance if changes have been reverted and then reapplied) just use the most recent revision. This may perform better, but may show incorrect revision numbers in File tab of the repository browser. |
The stop
command stops a running FishEye instance.
Options:
| Load configuration from the file at |
Usage:
fisheyectl fullscan [options] [repname ...]
The fullscan
command requests a full scan of the given repositories, or all repositories if no repository name is given.
Note: The fullscan command is only supported for CVS repositories.
Options:
| Load configuration from the file at |
Usage:
fisheyectl rescan [options] repname start end
Requests a rescan of the given repository between two specified revision ids.
Note: this operation is not supported by CVS repositories.
Options:
| Load configuration from the file at |
Usage:
fisheyectl reindex [options] [repname ...]
Requests a full reindex of the given repositories, or all repositories if no repository name is given.
Options:
| Load configuration from the file at |
The scannow
command requests an incremental scan from the command line.
Usage:
fisheyectl scannow -s [repname ...]
Options:
| Run synchronously; i.e. the command will not terminate until the requested indexing operation is complete. |
Repositories may be polled in the demand using this code:
fisheyectl.sh scannow [repname ...]
This will perform a one-time indexing operation.
Note: scannow can also be used on a repository which has a poll period.
Usage:
fisheyectl backup -f [filename]
The backup
command creates a zip archive containing important FishEye configuration files.
Options:
| Store the backup.zip to |