Bug 19705
Summary: | redials after disconnect | ||
---|---|---|---|
Product: | [Retired] Red Hat Linux | Reporter: | Need Real Name <gieldt> |
Component: | ppp | Assignee: | Thomas Woerner <twoerner> |
Status: | CLOSED NOTABUG | QA Contact: | Aaron Brown <abrown> |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | 7.0 | CC: | jkf, para_droid |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | i686 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2004-08-13 09:25:55 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
Need Real Name
2000-10-24 18:44:52 UTC
I also have this problem. I had to "downgrade" back to 6.2 to use my system. This is a serious problem in that it is imposible to use 7.0 if you need PPP. In my case I was able to get the "debug" button from the dial-up PPP suite to dial and hook up, but then I could not get FTP to work. (SEE Red Hat Service Request #160532, and ask Jeff about this.) Please increase the severity and priority of this bug. thanks, -vince vincentc (703)440-3692 For me, this problem does not occur if I disconnect using the small round disconnect button on the applet, it only occurs if I attempt to disconnect during the dialing process using the 'cancel' button. I think Vince is talking about a completely different bug, because a redialing modem hardly makes the system unusable (just pull the modem out of the socket, or kill the ppp process). Still, it was possible to halt the dialling process in 6.2, and it isnt in 7.0, and it ought to be fixed. I tried to take down ppp0 for security reasons today, but ppp0 was re- established automagically. I tried to take ppp0 down by typing ifdown ppp0 as root Further analysis showed that the ifdown-ppp and ifup-ppp scripts are not aligned: ifup-ppp checks on "${DEMAND}" != yes and ${DEMAND} = yes ifdown-ppp checks on "${DEMAND}" != no Examining ifcfg-ppp0 DEMAND is not present. This will lead to the behaviour described, when ifdown-ppp is changed to check on ${DEMAND} = yes everything works: therefore the following patch solves the problem: --- ifdown-ppp.redhatorig Sun Nov 12 22:33:16 2000 +++ ifdown-ppp Sun Nov 12 22:33:34 2000 @@ -7,7 +7,7 @@ CONFIG=$1 source_config -if [ "${DEMAND}" != no -a -f /var/run/ppp-${DEVICE}.pid ] ; then +if [ "${DEMAND}" = yes -a -f /var/run/ppp-${DEVICE}.pid ] ; then PID=`head -1 /var/run/ppp-${DEVICE}.pid` kill -TERM ${PID} sleep 2 I guess this one is solved Kind Regards, Martijn Kruithof I have the same problem. Apparently, the command "ifdown ppp0" does not terminate the ppp-watch program, which is monitoring and automatically restarts ppp if it sees the connection go down. My workaround is to do a ps, locate the ppp-watch program, and kill it. After killing ppp-watch, ifdown ppp0 will work properly. The problem is likely in the /etc/sysconfig/network-scripts/ifdown-ppp script; it should do the reverse of the ifup-ppp script in the same directory (but doesn't). Doug, did Martijn's patch not work for you? It just seems a little strange to hear you saying that ifdown-ppp doesnt work, when immediately above your comment is a patch which fixed it perfectly for me. Martijn's I did not notice Martijn's comment before posting my own (browser operator error -- failure to scroll comments completely before going to bottom to add own). It looks like it'll work. Another fix for those who prefer to wait for RedHat to service the problems (or who don't want the auto-restart at all)is to go into /etc/sysconfig/network-scripts and edit the ifcfg-ppp scripts to set PERSIST to no. As for me, I disabled the ppp scripts this morning because I just got DSL. Please verify this with a newer version of Red Hat Enterprise Linux or Fedora Core and reopen it against the new version if it still occurs. Closing as "not a bug" for now. |