Markku.Savela@tel.vtt.fi (VTT/TEL) (08/19/90)
Seems that for some reason the rgyd daemon dies on this machine once in a while. It's a minor annoyance only, except that sendmail (5.61++) stops working because it starts getting garbled user info? (Also logins take much longer etc..) Everytime this happens I cannot fix the problem from telnet session, because I cannot get to root (su says "no permission to su root"). What is exactly corrupted? And how to fix it that next time rgyd dies, I can su root and restart it from telnet session...? -- Markku Savela | savela@tel.vtt.fi Technical Research Centre of Finland | Telecommunications Laboratory | Markku.Savela@vtt.fi Otakaari 7 B, SF-02150 ESPOO, Finland | savela%vtttel@router.funet.fi
jnp@mjolner.tele.nokia.fi (J|rgen N|rgaard) (08/19/90)
In article <5067@hemuli.tik.vtt.fi> Markku.Savela@tel.vtt.fi (VTT/TEL) writes: Seems that for some reason the rgyd daemon dies on this machine once in a while. It's a minor annoyance only, except that sendmail (5.61++) stops working because it starts getting garbled user info? Almost; it's /bin/mail that fails to get certain information (e.g. mail group, varies a little though). Sigh! this but STILL around in 10.2 ?! (Also logins take much longer etc..) Everytime this happens I cannot fix the problem from telnet session, because I cannot get to root (su says "no permission to su root"). What is exactly corrupted? And how to fix it that next time rgyd dies, I can su root and restart it from telnet session...? add yourself to the "wheel" group. -- When is 10.3 due ? And will the registry/ncs etc. be more reliable ? Or the machine at least resort to the information keeps in /etc/{passwd,group} instead of stalling. -- ORIGIN '~jnp/stdDisclaimers'; [[--- Regards, J|rgen N|rgaard ('|' is '\o{}' in \LaTeX{}) | | e-mail: jnp@tele.nokia.fi || Telephone: <..>-358-0-511-5671 | | mail: Nokia Telecommunications, PL 33, SF-02601 Espoo Finland ---]]
hanche@imf.unit.no (Harald Hanche-Olsen) (08/19/90)
In article <JNP.90Aug19005444@mjolner.tele.nokia.fi> jnp@mjolner.tele.nokia.fi (J|rgen N|rgaard) writes: In article <5067@hemuli.tik.vtt.fi> Markku.Savela@tel.vtt.fi (VTT/TEL) writes: Everytime this happens I cannot fix the problem from telnet session, because I cannot get to root (su says "no permission to su root"). What is exactly corrupted? And how to fix it that next time rgyd dies, I can su root and restart it from telnet session...? add yourself to the "wheel" group. Evidently Markku is already in the wheel group, as he is usually able to su root. However, when rgyd dies group information tends to be unavailable, which is why he can't su root anymore. Only solutionn (I think) is to edit /etc/ttys to mark the ptys as secure, then do an ordinary login as root. This of course can be a slight security risk... A different question is, why does rgyd die? We had problems of this kind for a long time, and cured it by making sure that the node running the master rgyd also ran glbd. (This solution courtesy of many helpful people on the net).