Brocade switches have become one
of the most widely deployed components in most Storage Area Networks (SANs).
One thing that has led to Brocade’s success is their robust CLI, which allow
you to view and modify almost every aspect of their switch. This includes
zoning configurations, SNMP attributes, domain ids, switch names and network
addresses, etc. All of this configuration information is necessary for the
switch to function properly, and should be periodically backed up to allow speedy
recovery when disaster hits. Each Brocade switch comes with the “config Upload”
and “config Download” commands to back up a switch configuration to a remote
system or to restore a configuration from a remote system. Config Uploads has
two modes of operation: interactive mode and automatic mode. To use the
interactive mode to upload a config from a switch named switch1 to an ftp
server with the IP address 1.2.3.4, config Upload can be run to walk you
through backing up the configuration:
switch1:admin>
**configupload**
Server Name or IP Address [host]:
**1.2.3.4**
User Name [user]: **matty**
File Name [config.txt]:
**switch1_config.txt**
Protocol (RSHD or FTP) [rshd]:
**ftp**
Password:
Upload complete
After the configuration is
uploaded, you will have a text file with you switches configuration on the
remove server:
$ ls -l sw
-rw-r--r-- 1 matty other 7342 Jul
7 09:15 switch1_config.txt
To restore a configuration, you
can use the config Download command. Both of these commands allow the parameters
to be passed as arguments to the script, so they are ideal for automation
(there is a backup script on the Brocade support site that can be used to
automate configuration backups).
At exit () stage right. Or how my
Python program leaked file descriptors.
A year and a half ago I started
using Prometheus and grafana to graph metric data. This combination is
incredibly powerful and I have been absolutely blown away by the amount of
actionable intelligence I’ve been able to convey in our dashboards. Prometheus
has a number of exporters which can be used to retrieve metric data from remote
endpoints and stash it in its highly optimized time series database. There are
exporters for Mongo DB, Postgres, v Sphere, Nginx, HAProxy, JMX as well as
numerous other applications. If it’s a popular application there is most likely
an exporter for it. If an exporter isn’t available the Prometheus developers
have made it crazy easy to develop new ones with Python and Go.
I recently started playing with
pyvmomi which is a Python SDK for the VMware v Sphere API. The SDK allows you
to view and manipulate all aspects of v Sphere programmatically. This allows
you to automate things like collecting and graphing host and VM performance
data, adding disks, putting hosts into maintenance mode or generating common
reports. It’s powerful stuff! Being super curious I decided to write a
Prometheus VM performance metrics exporter so I could overlay business metrics
on top of HTTP responses which could then be laid on top of VM and host
performance metrics. After a couple hours of hacking my Flask-based exporter
was spitting out VM performance metrics. Or at least that’s what I thought.
Periodically the exporter would die and the following stack trace would be
written to the console:
1.2.3.4 - - [18/Mar/2018
01:10:48] "GET /metrics HTTP/1.1" 200 -
Traceback (most recent call last):
File "/usr/lib64/python2.7/SocketServer.py", line 295, in
_handle_request_noblock
self.process_request(request,
client_address)
File "/usr/lib64/python2.7/SocketServer.py", line 321, in
process_request
self.finish_request(request, client_address)
File "/usr/lib64/python2.7/SocketServer.py", line 334, in
finish_request
self.RequestHandlerClass(request,
client_address, self)
File "/usr/lib64/python2.7/SocketServer.py", line 651, in
__init__
self.finish()
File "/usr/lib64/python2.7/SocketServer.py", line 710, in
finish
self.wfile.close()
File "/usr/lib64/python2.7/socket.py", line 279, in close
self.flush()
File "/usr/lib64/python2.7/socket.py", line 303, in flush
self._sock.sendall(view[write_offset:write_offset+buffer_size])
My code was leaking file descriptors because
at exit () was leaving around references to the object that encapsulated the
file descriptor associated with the socket. Commenting out at exit () fixed my
issue and the code is now working splendidly. These types of issues pop up
during development and it’s incredibly fun debugging them! I learned a ton
about the Python debugger during this debugging session and became much more
familiar with the code for various Python modules. I also got to read the vast
majority of the PyVim and pyVmomi source code which made developing this a
snap! At exit (), stage right.
Some Services: shutdown unix command, brocade zoning commands, types of zoning in san, cisco zoning commands, vmax3 provisioning steps, snapvx, san zoning, unix flavors, timefinder snapvx, san storage explained, storage provisioning steps, emc snapvx, san zoning, symsg commands, how to backup brocade switch config, snapvx step by step, vmax snapvx, san zoning steps, gatekeeper device, zoning explained, emc vmax3, vsan vs zoning, dell emc vmax3, difference between vmax and vmax3, thin provisioning explained, data sovereignty requirements
Email Us: contact@sajidiqubal.com
Website: https://sajidiqubal.com/
