After bootup, (Run Level 5), I ran System->Administration->Services
I checked each service that was marked red (disabled). For a few of these disabled services, status was shown to be running. These services were:
akmods capi hsqldb isdn jetty lm_sensors racoon sandbox vmware zvbid
So, how would I prevent them from being started when they are already marked as disabled.
Also, several services which are enabled, the status was shown to be stopped. So, I had to start them. One of them was hal daemon. Others that I recall were: smbd, nmbd, ntpd, ...etc. I regret not saving their names on file to mention all of them.
Can anyone else confirm this when running Sysytem->Administration->Services?
On Fri, 2010-08-20 at 14:06 -0700, JD wrote:
After bootup, (Run Level 5), I ran System->Administration->Services
I checked each service that was marked red (disabled). For a few of these disabled services, status was shown to be running. These services were:
akmods capi hsqldb isdn jetty lm_sensors racoon sandbox vmware zvbid
So, how would I prevent them from being started when they are already marked as disabled.
Also, several services which are enabled, the status was shown to be stopped. So, I had to start them. One of them was hal daemon. Others that I recall were: smbd, nmbd, ntpd, ...etc. I regret not saving their names on file to mention all of them.
Can anyone else confirm this when running Sysytem->Administration->Services?
JD,
It's possible that system-config-services is not working correctly. As an alternative, use chkconfig to look at the default startup status for each of those services: # chkconfig --list | grep akmods akmods 0:off 1:off 2:on 3:on 4:on 5:on 6:off If you really want to disable akmods at startup: # chkconfig akmods off # service akmods stop Repeat for the other services.
HTH --Doc Savage Fairview Heights, IL
On 08/22/2010 01:38 PM, Robert G. (Doc) Savage wrote:
JD,
It's possible that system-config-services is not working correctly. As an alternative, use chkconfig to look at the default startup status for each of those services: # chkconfig --list | grep akmods akmods 0:off 1:off 2:on 3:on 4:on 5:on 6:off
$ chkconfig --list | grep akmods akmods 0:off 1:off 2:off 3:off 4:off 5:off 6:off
ditto with the rest: capi hsqldb isdn jetty lm_sensors racoon sandbox vmware zvbid
So, I still cannot explain this.
For example, capi is reported by system-config-services as running, BUT:
$ sudo service capi status capi not installed - Connection timed out (110)
So, how could it be reported as running.
akmod is also reported as running and ther is not S*akmods in any of /etc/rc[0123455].d
also, hsqldb is disabled, yet it is reported as running, same as isdn, jetty, lm_sensors, racoon, sandbox, vmware, zvbid
In summary, system-config-services cannot be relied on to correctly report what services are running or, disabled, or stopped.
On Sun, 2010-08-22 at 18:39 -0700, JD wrote:
On 08/22/2010 01:38 PM, Robert G. (Doc) Savage wrote:
JD,
It's possible that system-config-services is not working correctly. As an alternative, use chkconfig to look at the default startup status for each of those services: # chkconfig --list | grep akmods akmods 0:off 1:off 2:on 3:on 4:on 5:on 6:off
$ chkconfig --list | grep akmods akmods 0:off 1:off 2:off 3:off 4:off 5:off 6:off
ditto with the rest: capi hsqldb isdn jetty lm_sensors racoon sandbox vmware zvbid
So, I still cannot explain this.
For example, capi is reported by system-config-services as running, BUT:
$ sudo service capi status capi not installed - Connection timed out (110)
So, how could it be reported as running.
akmod is also reported as running and ther is not S*akmods in any of /etc/rc[0123455].d
also, hsqldb is disabled, yet it is reported as running, same as isdn, jetty, lm_sensors, racoon, sandbox, vmware, zvbid
In summary, system-config-services cannot be relied on to correctly report what services are running or, disabled, or stopped.
A service can be disabled (i.e. not enabled) in a run level, yet have been started by something else, perhaps inadvertently. Before declaring s-c-s to be buggy, it would be important to discover if these things actually *are* running or not, using ps, and if so what their parent processes are.
poc
On 08/22/2010 07:25 PM, Patrick O'Callaghan wrote:
On Sun, 2010-08-22 at 18:39 -0700, JD wrote:
On 08/22/2010 01:38 PM, Robert G. (Doc) Savage wrote:
JD,
It's possible that system-config-services is not working correctly. As an alternative, use chkconfig to look at the default startup status for each of those services: # chkconfig --list | grep akmods akmods 0:off 1:off 2:on 3:on 4:on 5:on 6:off
$ chkconfig --list | grep akmods akmods 0:off 1:off 2:off 3:off 4:off 5:off 6:off
ditto with the rest: capi hsqldb isdn jetty lm_sensors racoon sandbox vmware zvbid
So, I still cannot explain this.
For example, capi is reported by system-config-services as running, BUT:
$ sudo service capi status capi not installed - Connection timed out (110)
So, how could it be reported as running.
akmod is also reported as running and ther is not S*akmods in any of /etc/rc[0123455].d
also, hsqldb is disabled, yet it is reported as running, same as isdn, jetty, lm_sensors, racoon, sandbox, vmware, zvbid
In summary, system-config-services cannot be relied on to correctly report what services are running or, disabled, or stopped.
A service can be disabled (i.e. not enabled) in a run level, yet have been started by something else, perhaps inadvertently. Before declaring s-c-s to be buggy, it would be important to discover if these things actually *are* running or not, using ps, and if so what their parent processes are.
poc
Duh!!! Goes without saying! They are not running!
On Sun, 2010-08-22 at 19:33 -0700, JD wrote:
On 08/22/2010 07:25 PM, Patrick O'Callaghan wrote:
On Sun, 2010-08-22 at 18:39 -0700, JD wrote:
On 08/22/2010 01:38 PM, Robert G. (Doc) Savage wrote:
JD,
It's possible that system-config-services is not working correctly. As an alternative, use chkconfig to look at the default startup status for each of those services: # chkconfig --list | grep akmods akmods 0:off 1:off 2:on 3:on 4:on 5:on 6:off
$ chkconfig --list | grep akmods akmods 0:off 1:off 2:off 3:off 4:off 5:off 6:off
ditto with the rest: capi hsqldb isdn jetty lm_sensors racoon sandbox vmware zvbid
So, I still cannot explain this.
For example, capi is reported by system-config-services as running, BUT:
$ sudo service capi status capi not installed - Connection timed out (110)
So, how could it be reported as running.
akmod is also reported as running and ther is not S*akmods in any of /etc/rc[0123455].d
also, hsqldb is disabled, yet it is reported as running, same as isdn, jetty, lm_sensors, racoon, sandbox, vmware, zvbid
In summary, system-config-services cannot be relied on to correctly report what services are running or, disabled, or stopped.
A service can be disabled (i.e. not enabled) in a run level, yet have been started by something else, perhaps inadvertently. Before declaring s-c-s to be buggy, it would be important to discover if these things actually *are* running or not, using ps, and if so what their parent processes are.
poc
Duh!!! Goes without saying! They are not running!
Why does it go without saying? Nothing in your previous report indicates that you had checked the s-c-s version of reality using ps. Without that information, how can anyone decide where the problem really is?
poc
On 08/22/2010 08:34 PM, Patrick O'Callaghan wrote:
On Sun, 2010-08-22 at 19:33 -0700, JD wrote:
On 08/22/2010 07:25 PM, Patrick O'Callaghan wrote:
On Sun, 2010-08-22 at 18:39 -0700, JD wrote:
On 08/22/2010 01:38 PM, Robert G. (Doc) Savage wrote:
JD,
It's possible that system-config-services is not working correctly. As an alternative, use chkconfig to look at the default startup status for each of those services: # chkconfig --list | grep akmods akmods 0:off 1:off 2:on 3:on 4:on 5:on 6:off
$ chkconfig --list | grep akmods akmods 0:off 1:off 2:off 3:off 4:off 5:off 6:off
ditto with the rest: capi hsqldb isdn jetty lm_sensors racoon sandbox vmware zvbid
So, I still cannot explain this.
For example, capi is reported by system-config-services as running, BUT:
$ sudo service capi status capi not installed - Connection timed out (110)
So, how could it be reported as running.
akmod is also reported as running and ther is not S*akmods in any of /etc/rc[0123455].d
also, hsqldb is disabled, yet it is reported as running, same as isdn, jetty, lm_sensors, racoon, sandbox, vmware, zvbid
In summary, system-config-services cannot be relied on to correctly report what services are running or, disabled, or stopped.
A service can be disabled (i.e. not enabled) in a run level, yet have been started by something else, perhaps inadvertently. Before declaring s-c-s to be buggy, it would be important to discover if these things actually *are* running or not, using ps, and if so what their parent processes are.
poc
Duh!!! Goes without saying! They are not running!
Why does it go without saying? Nothing in your previous report indicates that you had checked the s-c-s version of reality using ps. Without that information, how can anyone decide where the problem really is?
poc
Please, stop the noise! I am sure there are people who will be tryinf this for themselves and see that it is the case. There is a problem with system-config-services. Enough from you.
On Sun, 2010-08-22 at 20:52 -0700, JD wrote:
There is a problem with system-config-services.
JD,
I would have to agree. Time to BZ s-c-s. Fortunately chkconfig and ps give you independent confirmations that the unwanted services are not actually running, no matter what s-c-s may say.
Good luck. --Doc Savage Fairview Heights, IL
On 08/22/2010 09:16 PM, Robert G. (Doc) Savage wrote:
On Sun, 2010-08-22 at 20:52 -0700, JD wrote:
There is a problem with system-config-services.
JD,
I would have to agree. Time to BZ s-c-s. Fortunately chkconfig and ps give you independent confirmations that the unwanted services are not actually running, no matter what s-c-s may say.
Which is fine, if you want to do these extra steps to get this sort of confirmation. But the whole purpose of this util is to tell you at a glance. And it is giving contradictory info. If a servce is disabled, and I have nothing that starts it, then the sys config services should not say the status is running. In fact, even clicking on Stop, the status remains 'running'. Also, I am not sure if this bug should be filed against Fedora, as it is present in other distros.
On Sun, 2010-08-22 at 21:53 -0700, JD wrote:
On 08/22/2010 09:16 PM, Robert G. (Doc) Savage wrote:
On Sun, 2010-08-22 at 20:52 -0700, JD wrote:
There is a problem with system-config-services.
JD,
I would have to agree. Time to BZ s-c-s. Fortunately chkconfig and ps give you independent confirmations that the unwanted services are not actually running, no matter what s-c-s may say.
Which is fine, if you want to do these extra steps to get this sort of confirmation. But the whole purpose of this util is to tell you at a glance. And it is giving contradictory info. If a servce is disabled, and I have nothing that starts it, then the sys config services should not say the status is running. In fact, even clicking on Stop, the status remains 'running'. Also, I am not sure if this bug should be filed against Fedora, as it is present in other distros.
JD,
Go ahead and submit a Bugzilla report against s-c-s citing exactly what you are seeing, what it's not doing, etc. It's been my experience that the service-config-whatever scripts often get lost in the shuffle of new packages. When their maintainers get good descriptive Bugzilla reports, they can frequently push out a good fix in short order.
For example, about six months ago a problem with system-config-httpd was fixed after lying dormant for almost two releases. IIRC that problem turned out to be in alchemy, used by s-c-h. It took the maintainer a couple of months to find and fix the problem, but he eventually got it.
Good luck. --Doc Savage Fairview Heights, IL
On 08/23/2010 08:15 AM, Robert G. (Doc) Savage wrote:
On Sun, 2010-08-22 at 21:53 -0700, JD wrote:
On 08/22/2010 09:16 PM, Robert G. (Doc) Savage wrote:
On Sun, 2010-08-22 at 20:52 -0700, JD wrote:
There is a problem with system-config-services.
JD,
I would have to agree. Time to BZ s-c-s. Fortunately chkconfig and ps give you independent confirmations that the unwanted services are not actually running, no matter what s-c-s may say.
Which is fine, if you want to do these extra steps to get this sort of confirmation. But the whole purpose of this util is to tell you at a glance. And it is giving contradictory info. If a servce is disabled, and I have nothing that starts it, then the sys config services should not say the status is running. In fact, even clicking on Stop, the status remains 'running'. Also, I am not sure if this bug should be filed against Fedora, as it is present in other distros.
JD,
Go ahead and submit a Bugzilla report against s-c-s citing exactly what you are seeing, what it's not doing, etc. It's been my experience that the service-config-whatever scripts often get lost in the shuffle of new packages. When their maintainers get good descriptive Bugzilla reports, they can frequently push out a good fix in short order.
For example, about six months ago a problem with system-config-httpd was fixed after lying dormant for almost two releases. IIRC that problem turned out to be in alchemy, used by s-c-h. It took the maintainer a couple of months to find and fix the problem, but he eventually got it.
Good luck. --Doc Savage Fairview Heights, IL
OK. Sounds reasonable. I am wondering whether or not to open the bug in Fedora bugzilla or the more general Redhat Bugzilla? Is there a feedback loop between fedora bugzilla and redhat bugzilla?
On Mon, 2010-08-23 at 09:22 -0700, JD wrote:
On 08/23/2010 08:15 AM, Robert G. (Doc) Savage wrote:
On Sun, 2010-08-22 at 21:53 -0700, JD wrote:
On 08/22/2010 09:16 PM, Robert G. (Doc) Savage wrote:
On Sun, 2010-08-22 at 20:52 -0700, JD wrote:
There is a problem with system-config-services.
JD,
I would have to agree. Time to BZ s-c-s. Fortunately chkconfig and ps give you independent confirmations that the unwanted services are not actually running, no matter what s-c-s may say.
Which is fine, if you want to do these extra steps to get this sort of confirmation. But the whole purpose of this util is to tell you at a glance. And it is giving contradictory info. If a servce is disabled, and I have nothing that starts it, then the sys config services should not say the status is running. In fact, even clicking on Stop, the status remains 'running'. Also, I am not sure if this bug should be filed against Fedora, as it is present in other distros.
JD,
Go ahead and submit a Bugzilla report against s-c-s citing exactly what you are seeing, what it's not doing, etc. It's been my experience that the service-config-whatever scripts often get lost in the shuffle of new packages. When their maintainers get good descriptive Bugzilla reports, they can frequently push out a good fix in short order.
For example, about six months ago a problem with system-config-httpd was fixed after lying dormant for almost two releases. IIRC that problem turned out to be in alchemy, used by s-c-h. It took the maintainer a couple of months to find and fix the problem, but he eventually got it.
Good luck. --Doc Savage Fairview Heights, IL
OK. Sounds reasonable. I am wondering whether or not to open the bug in Fedora bugzilla or the more general Redhat Bugzilla? Is there a feedback loop between fedora bugzilla and redhat bugzilla?
JD,
I'd submit it under F13 s-c-s and let the maintainers decide where to apply the fix.
--Doc Savage Fairview Heights, IL
On Mon, 23 Aug 2010 09:22:34 -0700, JD wrote:
I am wondering whether or not to open the bug in Fedora bugzilla or the more general Redhat Bugzilla?
What do you mean? Both use http://bugzilla.redhat.com
You could jump in at http://bugz.fedoraproject.org/system-config-services out of convenience, though.
Is there a feedback loop between fedora bugzilla and redhat bugzilla?
? Obsolete question I guess, given what I replied above.
On Sun, 2010-08-22 at 20:52 -0700, JD wrote:
Please, stop the noise! I am sure there are people who will be tryinf this for themselves and see that it is the case. There is a problem with system-config-services. Enough from you.
If this is how you are going to treat people who try to help you, then nobody is going to want to help.
On Sun, 2010-08-22 at 20:52 -0700, JD wrote:
Why does it go without saying? Nothing in your previous report
indicates
that you had checked the s-c-s version of reality using ps. Without
that
information, how can anyone decide where the problem really is?
poc
Please, stop the noise! I am sure there are people who will be tryinf this for themselves and see that it is the case. There is a problem with system-config-services. Enough from you.
Unbelievable. This is your response to someone asking for further information? Good luck with Bugzilla.
poc
-
Greg Woods -
JD -
Michael Schwendt -
Patrick O'Callaghan -
Robert G. (Doc) Savage