You can disregard this patch if you find it horrible.
I haven't thought about something that would fit better into
our logging system yet.
---
gui.py | 2 ++
network.py | 24 ++++++++++++++++++++++++
2 files changed, 26 insertions(+), 0 deletions(-)

diff --git a/gui.py b/gui.py
index 307f5dc..6d7ecc9 100755
--- a/gui.py
+++ b/gui.py
@@ -976,7 +976,9 @@ class InstallInterface(InstallInterfaceBase):
if not just_setup:
self.anaconda.network.updateActiveDevices([install_device])

+ network.logIfcfgFiles(header="========== before nm-c-e run
")
runNMCE(self.anaconda)
+ network.logIfcfgFiles(header="========== after nm-c-e run
")

self.anaconda.network.update()

diff --git a/network.py b/network.py
index 9bbb935..a7c474e 100644
--- a/network.py
+++ b/network.py
@@ -238,6 +238,30 @@ def getActiveNetDevs():
ret.sort()
return ret

+def logIfcfgFile(path, header="
"):
+ logfile = ifcfgLogFile
+ try:
+ f = open(path, 'r')
+ except OSError as e:
+ log.warning("Exception while opening file %s: %s" % (path, str(e)))
+ return
+ lf = open(logfile, 'a')
+ lf.write(header)
+ lf.write(f.read())
+ lf.close()
+ f.close()
+
+def logIfcfgFiles(header="
"):
+
+ lf = open(ifcfgLogFile, 'a')
+ lf.write(header)
+ lf.close()
+
+ devprops = isys.getDeviceProperties(dev=None)
+ for device in devprops:
+ path = "%s/ifcfg-%s" % (netscriptsDir, device)
+ logIfcfgFile(path, "===== %s
" % (path,))
+
class NetworkDevice(IfcfgFile):

def __init__(self, dir, iface, logfile='/tmp/ifcfg.log'):
--
1.6.0.6

_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Not horrible, but I'm not crazy about the spurious log file output such as:

========== before nm-c-e run

Those entries can stay, but make them log at the DEBUG level.

On Tue, 27 Apr 2010, Radek Vykydal wrote:


You can disregard this patch if you find it horrible.
I haven't thought about something that would fit better into
our logging system yet.
---
gui.py | 2 ++
network.py | 24 ++++++++++++++++++++++++
2 files changed, 26 insertions(+), 0 deletions(-)

diff --git a/gui.py b/gui.py
index 307f5dc..6d7ecc9 100755
--- a/gui.py
+++ b/gui.py
@@ -976,7 +976,9 @@ class InstallInterface(InstallInterfaceBase):
if not just_setup:
self.anaconda.network.updateActiveDevices([install_device])

+ network.logIfcfgFiles(header="========== before nm-c-e run
")
runNMCE(self.anaconda)
+ network.logIfcfgFiles(header="========== after nm-c-e run
")

self.anaconda.network.update()

diff --git a/network.py b/network.py
index 9bbb935..a7c474e 100644
--- a/network.py
+++ b/network.py
@@ -238,6 +238,30 @@ def getActiveNetDevs():
ret.sort()
return ret

+def logIfcfgFile(path, header="
"):
+ logfile = ifcfgLogFile
+ try:
+ f = open(path, 'r')
+ except OSError as e:
+ log.warning("Exception while opening file %s: %s" % (path, str(e)))
+ return
+ lf = open(logfile, 'a')
+ lf.write(header)
+ lf.write(f.read())
+ lf.close()
+ f.close()
+
+def logIfcfgFiles(header="
"):
+
+ lf = open(ifcfgLogFile, 'a')
+ lf.write(header)
+ lf.close()
+
+ devprops = isys.getDeviceProperties(dev=None)
+ for device in devprops:
+ path = "%s/ifcfg-%s" % (netscriptsDir, device)
+ logIfcfgFile(path, "===== %s
" % (path,))
+
class NetworkDevice(IfcfgFile):

def __init__(self, dir, iface, logfile='/tmp/ifcfg.log'):



- --
David Cantrell <dcantrell@redhat.com>

Red Hat / Honolulu, HI

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEARECAAYFAkvfjpIACgkQ5hsjjIy1Vkm56gCg6lvo6KlBz9 aP4d1tGi9y8yy4
l5wAoOnwwytD/qouvqnbyq4X5BX7SXUV
=qBsy
-----END PGP SIGNATURE-----

_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list

David Cantrell wrote:

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Not horrible, but I'm not crazy about the spurious log file output
such as:


========== before nm-c-e run

Those entries can stay, but make them log at the DEBUG level.


It is logged into separate /tmp/ifcfg.log, so it doesn't polute
log files we have. Also I don't think it is using much resources,
and it can be pretty handy info that can be asked from bz reporter,
it is a source of information about what user set in nm-c-e,
so I'd not limit it to debug mode.

Radek

_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wed, 5 May 2010, Radek Vykydal wrote:


David Cantrell wrote:

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Not horrible, but I'm not crazy about the spurious log file output such as:

========== before nm-c-e run

Those entries can stay, but make them log at the DEBUG level.


It is logged into separate /tmp/ifcfg.log, so it doesn't polute
log files we have. Also I don't think it is using much resources,
and it can be pretty handy info that can be asked from bz reporter,
it is a source of information about what user set in nm-c-e,
so I'd not limit it to debug mode.


OK, that's fine then.

- --
David Cantrell <dcantrell@redhat.com>

Red Hat / Honolulu, HI

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEARECAAYFAkvi1z8ACgkQ5hsjjIy1Vkn46QCgjMFeduATjY ExaTxeE5Is5iVU
qQgAn0m+erZnKYg+A1W1q/61go9HVVgo
=L2WW
-----END PGP SIGNATURE-----

_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list