James-san, >> Q2. They want to issue TA command after port is actually opened, >> but is there any way to do it via script like chat script? > > Yes, there are at least two ways to do that, and they're described in > the pppd(1M) man page.
If it's possible to do it via chat script, please let us know how they need to describe it. If it's not supported way, it's OK. Best regards, Yozo Nitta James Carlson wrote: > Yozo Nitta writes: > >> I got additional question from CU. > > > It would help *greatly* if you could get a statement from the customer > about what problem the user is seeing. I don't mind answering > questions, but answering these detailed questions about system > behavior seems like a very inefficient way to solve a problem. It > seems to me as though we're going in circles around the man pages. > >> Q1. Could you tell us how time until port is opened is decided? >> Does it depend on TA or hardware on server? > > > It depends on the application behavior. When the application makes a > call to the open(2) function, then that action opens the port and > causes DTR (ER) to be asserted. > > In the case of pppd(1M), the port is normally opened immediately when > /usr/bin/pppd is invoked. The time when /usr/bin/pppd is invoked is > under user control -- usually either via a script, some other daemon > activity (such as ttymon or mgetty), or direct user invocation. > > The exceptions to that rule are: > > - When the pppd "demand" option is used, pppd will create the spppN > IPv4/IPv6 interfaces, but will not open the serial port until > packets are seen on those interfaces. When the system attempts to > transmit a packet, pppd will open the serial port and begin the > process of bringing up the link. > > - After completing a call with the "persist" option, the user may > specify a "holdoff" period. If one is specified, then that timer > is allowed to elapse before the port is reopened. > >> Q2. They want to issue TA command after port is actually opened, >> but is there any way to do it via script like chat script? > > > Yes, there are at least two ways to do that, and they're described in > the pppd(1M) man page. > > connect script Use the executable or shell command > specified by script to set up the > serial line. This script would typi- > cally use the chat(1M) program to > dial the modem and start the remote > PPP session. A value for this option > originating from a privileged source > cannot be overridden by a non- > privileged user. > > init script Run the executable or shell command > specified by script to initialize the > serial line. This script would typi- > cally use the chat(1M) program to > configure the modem to enable auto- > answer. A value for this option from > a privileged source cannot be over- > ridden by a non-privileged user. > > There's also a "welcome" script that would function in this way, but I > don't think it's applicable for this scenario. > _______________________________________________ networking-discuss mailing list [email protected]
