Bug#220025: [Pkg-sysvinit-devel] Bug#220025: Will fix
Petter Reinholdtsen wrote: > Why do you want to stop it at the end of runlevel S, instead of > stopping it at the start of runlevel 1 (single user)? I would suspect > adding for example 1:K99bootlogd would solve this problem, as it would > stop bootlogd just before starting S20single. On boot into single-user mode the system never leaves runlevel S. The way we tend to think of runlevels in Debian is: S: Transient mode for one-time-only system initialization 1: Single-user mode -- no services running 2-5: Multi-user modes However, historically the runlevels seem to have been conceived this way: S: Initial, single-user mode 1: Transient mode for one-time-only return to single-user mode 2-5: Multi-user modes I prefer to think of things in the Debian way, because the runlevel directory rc1.d is much more analogous to rc[2-5].d than it is to rcS.d. However, it remains a fact that when booting into single-user mode and when switching to single-user mode via runlevel 1, the end result is that the system is in runlevel "S". -- Thomas -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#220025: Will fix
[Thomas Hood] > At the end of the initscript sequence for runlevel S we need to know > whether or not we are going to switch to a multi-user runlevel (2 > through 5). (If we are not going to switch to a multi-user runlevel > then we must stop bootlogd; otherwise we leave bootlogd running and > it will be stopped at the end of the initscript sequence for the > multi-user runlevel.) How can this information be obtained from > runlevel(8)? Why do you want to stop it at the end of runlevel S, instead of stopping it at the start of runlevel 1 (single user)? I would suspect adding for example 1:K99bootlogd would solve this problem, as it would stop bootlogd just before starting S20single. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#220025: Will fix
Thomas Hood ([EMAIL PROTECTED]) wrote: > I propose to fix this bug by adding a script at S:S99 that stops bootlogd > if the system is being booted with the "single" parameter. Anyone see any > problem with this idea? There are other "aliases" for single, such as "1". Some people might come up in single user mode using one of those alternative boot parameters. I don't know whether there's an exhaustive list of all of them anywhere, either. I suggest using the runlevel(8) command instead, assuming that it returns reliable information. The man page says it's safe to use in rc scripts, but I haven't tried it myself. -- Greg Wooledge | "Truth belongs to everybody." [EMAIL PROTECTED] |- The Red Hot Chili Peppers http://wooledge.org/~greg/ | signature.asc Description: Digital signature
Bug#220025: Will fix
I propose to fix this bug by adding a script at S:S99 that stops bootlogd if the system is being booted with the "single" parameter. Anyone see any problem with this idea? -- Thomas Hood -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
