}From: inf_bb@hermes1.sps.mot.com (Bob Baskett)
}Subject: tbstat -u question
}Date: Mon, 22 Nov 1993 14:29:08 -0700 (MST)
}X-Informix-List-Id: <list.3129>
}can anyone tell me what the pid 0 process is in the following report. what
}happened was a tbmonitor-mode-immediate shut down was performed while someone
}else was doing a tbmonitor. when will this process go away?
}
}RSAM Version 5.00.UC3 -- On-Line -- Up 04:57:34 -- 54536 Kbytes
}
}Users
}address flags pid user tty wait tout locks nreads nwrites
}10001ba8 ------D 177 root console 0 0 0 3431 11817
}10001c18 ------D 0 root console 0 0 0 0 0
}10001c88 ------F 178 root 0 0 0 0 0
This process 0 is an non-existent tbundo process. This slot is reserved
for tbundo. Tbundo is another incarnation of tbinit (like tbpgcl is
the page cleaner incarnation of tbinit), and it comes into existence
(non-zero pid) when an sqlturbo dies unexpectedly inside a transaction.
Basically, it rolls back the transaction on behalf of the dead sqlturbo.
This slot, therefore, normally has a zero pid. In general, you don't want
the zero pid to go away -- if it goes away, it is replaced by an actual
process.
Yours,
Jonathan Leffler (johnl@informix.com) #include <disclaimer.h>