Age | Commit message (Collapse) | Author |
|
|
|
from shell scripts.
|
|
|
|
perl only allows a require to be done once. If it is done in two
included modules the module or function in the module used second
doesn't have the data in the required file in its scope. To prevent
this I created a third module that contains the functions that require
'sys/ioctl.ph' so that there is no longer a conflict.
|
|
successful (complete) commits for all commit methods.
|
|
to manage chunks. Command interpreter is now in vyatta-cfg-run and common functions have been moved to vyatta-op
|
|
Also rotate the config related log files under /var/log/vyatta so that
they don't fill up the root filesystem.
|
|
* do NOT reinvent the wheel (directory traversal, sorting, script execution, etc.) when the functionality is already provided by the system.
* eliminate hard-coded hook directory in various feastures. everyone should obtain the directory path from the config backend.
|
|
(cherry picked from commit 1b2a0fd1ae1e6dfc18e4f75f73cd7befb47cf538)
|
|
|
|
|
|
|
|
* extend output/assert mechanism and simplify code.
* when handling internal error, automatically detect perl context and croak if the library is used from perl.
|
|
Perl versions of commonly used backquote escapes.
|
|
* add new boot-time config loading script that uses shell API to access backend directly, bypassing wrapper.
* use new loadFile implementation in the backend to perform the actual "load" operation (without "commit"). this reduces the "pre-commit" load time by ~60%.
|
|
|
|
The scripts in /etc/bash_completion.d are not named with number
like init files. Also completion script should not be marked executable.
|
|
Part of earlier effort to use less sudo. Not currently used.
|
|
* add parser to package build.
* add prefix, error handling, etc.
* fix handling of empty and 1-char values in lex.
* add API function.
|
|
* change showCfg to use new code.
* remove old files.
|
|
* separate data from algorithm.
* prepare for unified input/output framework.
|
|
|
|
|
|
* replace the original ConfigOutput perl module.
* simplify logic and fix bugs in original code.
|
|
* unify node sorting implementation into the backend library.
* allow future implementation of per-node, customized sorting policy.
|
|
|
|
|
|
|
|
|
|
|
|
The entitlement bug can be fixed https://bugzilla.vyatta.com/show_bug.cgi?id=5354 by adding a hook into this directory.
|
|
remove these comments from the comments in the action configuration the user should:
1) save their current configuration
2) run this script on the saved configuration
3) reload their configuration (which will strip the migration comments from the active configuration tree)
|
|
Less chance of name collision, and useful for /proc as well.
|
|
|
|
By using a command (instead of echo), it is possible to set capability
and avoid sudo.
|
|
* use "enumeration" template field for both CLI completion and GUI2
enumeration support.
* CLI completion now uses enumeration if it exists.
* add "existing-interfaces" script as an example.
|
|
/* to create a comment */
> comment [PATH-TO-NODE] "my comments"
/* to delete a comment */
> comment [PATH-TO-NODE] ""
|
|
supported.
|
|
It is really part of interfaces not vyatta-cfg
|
|
Bug 4386
vyatta-ofr is out of date product name.
|
|
This reverts commit ba1376cd853e6b48cd3c6d32a71e5bb6672c3ff9.
Conflicts:
Makefile.am
|
|
The files in modprobe.d are not supposed to be marked as executable.
Fixes lintian warning.
|
|
This is a known upstream kernel bug, but turning off DMA should
solve majority of problem.
|
|
Future versions of modutils ignore files not named .conf, and latest Debian
tools warn about that. Fix it before the bug happens.
|
|
|
|
All the other templates and scripts related to interfaces
are in vyatta-cfg-system
|
|
|
|
This change implements a mechanism for setting the IRQ affinity of NIC
interrupts in a way that is optimally balanced. The affinity settings
are static, and are set up at configuration time. I've added "auto"
as a new legal value to the "interfaces ethernet ethX smp_affinity"
configuration parameter. If this value is set, then the system
attempts to perform automatic IRQ affinity assignment.
At this time, automatic IRQ affinity is supported only for multiqueue
NICs that use the igb or bnx2 drivers. Even with these two,
differences in the organization of queues and their naming convention
necessitate some NIC-specific code. The assignment strategy attempts
to assign all queues of all NICs that share the same queue number to
the same CPU. This has been shown to be optimal for IP forwarding
workloads in 2.6.31 kernels.
Setting the value to "auto" for NICs that use other drivers will be a
no-op.
|
|
* convert subshell invocations to library function calls.
* reduce boot time by ~40 seconds for "certain configuration".
|
|
corresponding changes made to node.defs
|