I don't think it is a memory leak within the load balancer. elb does not
endlessly add to the host queue, see the first few lines of parse_qhost:
def parse_qhost(self, string):
"""
this function parses qhost -xml output and makes a neat array
takes in a string, so we can pipe in output from ssh.exec('qhost -xml')
"""
self.hosts = [] # clear the old hosts
https://github.com/jtriley/StarCluster/blob/develop/starcluster/balancers/sge/__init__.py
Looks to be an XML parser issue. Maybe ELB is using the parser incorrectly.
How truly huge is the qhosts -xml output? I would think it would be about 1
xml record per host in the cluster, and the XML output contains many many
parameters about the status of the host.
Received on Wed Jan 11 2012 - 18:30:00 EST
This archive was generated by
hypermail 2.3.0.