[Stackoverflow] [Progress OpenEdge ABL] Lighttpd and CGI Timeout issue

[Raymond Herring]

Recently updated from debian 8 to debian 10, which meant Lighttpd was upgraded from 1.4.35 to 1.4.53, but ever since the upgrade, i've been seeing errors on our database server which points to the original client no longer being reachable.

Our database server runs Progress Openedge, and we use Progress's messenger service called 'Webspeed'.

I've checked online and found references to the error in question in the logs, but it pertains to IIS 4 being upgraded to newer versions, where the newer versions have a lower timeout for CGI.

We use mod_cgi with lighttpd, as it's the only thing that will work (Can't use FastCGI in this scenario).

Prior to the upgrade, my lighttpd config was working 100%, after the upgrade, the config is the same, but it's been throwing up these errors.

The error in the Webspeed logfile is: Progress KB - Low CGI Timeout value causes WebWrite errors (errors 6404) within WebSpeed server log

Lighttpd config:

server.modules = (
        "mod_access",
        "mod_accesslog",
        "mod_alias",
        "mod_compress",
        "mod_expire",
        "mod_redirect",
        "mod_setenv",
)

server.upload-dirs          = ( "/var/cache/lighttpd/uploads" )
server.errorlog             = "/var/log/lighttpd/error.log"
server.pid-file             = "/var/run/lighttpd.pid"
server.username             = "www-data"
server.groupname            = "www-data"
server.port                 = 80

server.document-root = "/var/www"
accesslog.filename   = "/var/log/lighttpd/access.log"

# Keep-Alive stuff
# Was at 32 and 3
server.max-keep-alive-requests = 0
server.max-keep-alive-idle     = 3


index-file.names            = ( "index.php", "index.html", "index.lighttpd.html" )
url.access-deny             = ( "~", ".inc" )
static-file.exclude-extensions = ( ".php", ".pl", ".fcgi" )

include_shell "/usr/share/lighttpd/use-ipv6.pl " + server.port
include_shell "/usr/share/lighttpd/create-mime.assign.pl"
include_shell "/usr/share/lighttpd/include-conf-enabled.pl"

include "myconf.conf"

myconf.conf is just general stuff plus some supposed performance tuning stuff

server.modules += ( "mod_status" )

server.max-fds       = 2048                #Normally 1024? 
server.event-handler = "linux-sysepoll"    #Normally level-triggered
server.use-noatime = "enable"              #Normally disabled

$HTTP["remoteip"] == "nnn.nnn.nnn.0/24" {
  status.status-url = "/server-status"
}

$HTTP["host"] == "www.example.com.au" {
  server.document-root = "/var/www"
}

alias.url += ("/pse" => "/var/www/static/" )
alias.url += ("/assets" => "/var/www/assets/dist")

As said, nothing has changed config wise from when I was running Debian 8.

Looking online, best I can find with regards to lighttpd config is a server.max-read-idle and server.max-write-idle, but the documentation doesn't really go into much detail about them.

I can do some fine tuning on the database server itself, but that would just be to change the retry count from the default of 10 to 0 so that if the client vanishes, the Webspeed messenger doesn't bother retrying for 10 times before giving up, should probably do that anyway, but would like a lighttpd fix too.

Continue reading...