Be the first to post a text review of finetd, an Internet Superserver. Rate and review a project by clicking thumbs up or thumbs down in the right column.

• Code committed

  kjak committed revision 27 to the finetd, an Internet Superserver SVN repository, changing 9 files

  posted by kjak 129 days ago

• Code committed

  kjak committed revision 26 to the finetd, an Internet Superserver SVN repository, changing 1 files

  posted by kjak 129 days ago

• Code committed

  kjak committed revision 25 to the finetd, an Internet Superserver SVN repository, changing 16 files

  posted by kjak 130 days ago

• Code committed

  kjak committed revision 23 to the finetd, an Internet Superserver SVN repository, changing 1 files

  posted by kjak 130 days ago

• Code committed

  kjak committed revision 22 to the finetd, an Internet Superserver SVN repository, changing 13 files

  posted by kjak 130 days ago

• Code committed

  kjak committed revision 21 to the finetd, an Internet Superserver SVN repository, changing 6 files

  posted by kjak 130 days ago

• File released: /finetd/0.2/finetd-0.2.tar.gz

  posted 625 days ago

• File released: /finetd/0.2/finetd-0.2.tar.bz2

  posted 625 days ago

• finetd 0.2 file released: finetd-0.2.tar.gz

  o Now if a spawned process fails to exec() the specified command, the child exits with abort() so finetd knows something is wrong. Right now I'm going to wait until a second abort before disposing of it and taking the socket off of the queue, but I may just change this to one later. It's entirely possible the executed process can exit abnormally in this manner rather than my abort() call, but in that case it's probably best to take it off the list anyway. o Now a child process is actually checked for termination rather than just any status change (such as stopping) o Fixed inetd.conf parsing for seqpacket o Allow longer protocol names in inetd.conf field o Now finetd exits if chown() or chmod() fail against a UNIX domain socket as this can pose a security risk o Now it's tested whether or not a *datagram* socket is used with the 'nowait' inetd.conf field rather than just UDP sockets o Added some additional sanity checks against inetd.conf entries, like an attempt to use TCP with a non-stream socket for example o Added a couple #includes to fix compilation on FreeBSD (and probably other systems as well). They are the proper headers to include, but my Linux box with glibc didn't need them so I didn't catch the omission. o Added more detail to some error messages

  posted 626 days ago

• finetd 0.2 file released: finetd-0.2.tar.bz2

  o Now if a spawned process fails to exec() the specified command, the child exits with abort() so finetd knows something is wrong. Right now I'm going to wait until a second abort before disposing of it and taking the socket off of the queue, but I may just change this to one later. It's entirely possible the executed process can exit abnormally in this manner rather than my abort() call, but in that case it's probably best to take it off the list anyway. o Now a child process is actually checked for termination rather than just any status change (such as stopping) o Fixed inetd.conf parsing for seqpacket o Allow longer protocol names in inetd.conf field o Now finetd exits if chown() or chmod() fail against a UNIX domain socket as this can pose a security risk o Now it's tested whether or not a *datagram* socket is used with the 'nowait' inetd.conf field rather than just UDP sockets o Added some additional sanity checks against inetd.conf entries, like an attempt to use TCP with a non-stream socket for example o Added a couple #includes to fix compilation on FreeBSD (and probably other systems as well). They are the proper headers to include, but my Linux box with glibc didn't need them so I didn't catch the omission. o Added more detail to some error messages

  posted 626 days ago