SALT-SSH: AttributeError: 'module' object has no attribute 'fromstringlist' 

This error triggers when the salt-master is CentOS6 and the target is CentOS7. 
The salt-master on CentOS6 would copy ElementTree.py to the minion on CentOS7 but 
python2.6's lib cannot run directly on python2.7.


To overcome this issue,

On the salt-master:

1). Find the thin.py

# find / -name thin.py
/usr/lib/python2.6/site-packages/salt/runners/thin.py
/usr/lib/python2.6/site-packages/salt/utils/thin.py

In /usr/lib/python2.6/site-packages/salt/utils/thin.py

try:
    import xml
    HAS_XML = True
except ImportError:
    HAS_XML = False

Modify as:

try:
    import xml
    HAS_XML = False
except ImportError:
    HAS_XML = False


2). remove the thin.tgz from cache

# rm /var/cache/salt/master/thin/thin.tgz

3). Perform the salt-ssh command again with -W
( -W for Select a random temp dir to deploy on the remote system. The dir will be cleaned
after the execution.)

# salt-ssh -W rnd test.ping

# salt-ssh -W rnd disk.usage

Comments

Post a Comment

Popular posts from this blog

Resolved: DOCKER: Error response from daemon: Could not attach to network / rpc error: code = 7 desc = network not manually attachable.

Image
I got this error: 'docker: Error response from daemon: Could not attach to network <nw-name>: rpc error: code = 7 desc = network <nw-name> not manually attachable.' when i try to create the container on the overlay network on swarm mode enabled, In docker 1.12 with swarm mode enabled, "Swarm" scoped networks (see the "SCOPE" column) cannot be used for docker run, only for services and now it has been rectified. I'm using updated version of docker, but still why i'm experiencing the error ? because the network just created but not manually attachable. To fix the issue we gonna use --attachable flag while creating network. [ --attachable             Enable manual container attachment ] (Need to create service on the overlay network, so that the overlay network will available / list on the worker node) docker inspect attnw (attnw is our test network)should show your own c
Read more

yum failed 6 times. Cannot continue!

Error while cPanel installation: (ERROR): - ssystem [EXIT_CODE] '/usr/bin/yum' exited with 1 (ignored) (DEBUG): - ssystem [END] (FATAL): yum failed 6 times. Cannot continue! fix: It looks like you have some name resolution issues on the server. Just run the following commands on your server and it will resolve the name resolution issues. #echo "nameserver 8.8.8.8" >> /etc/resolv.conf #echo "nameserver 8.8.4.4" >> /etc/resolv.conf if it is not working properly still facing same error, then here we go, # rpm -e --nodeps bind-chroot # rpm -e --nodeps bind # yum install bind
Read more

unexpectedly shrunk window (repaired) in dmesg log - TCP Peer

Don't Panic, This normally occurs when a client decides to reduce its TCP window size, without the server expecting it. This can be the case when fragmentation is an issue, or when the client is using an embedded device with very little NIC buffer memory. This is a completely normal behaviour, and you’re likely to see quite a few such packets in your log. The messages are informational only, and are used to debug networking issues. I’d be worried if you saw hundreds of thousands of these packets, since there are attacks that involve packet fragmentation and small window sizes, but otherwise it’s just the normal sort of noise you should expect to see on any internet-facing network. In fact, the “repaired” part of your message is showing that your network driver fixed the issue, which is usually done by concatenating the payloads of two fragmented packets together. Shouldn’t be an issue at all. … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … 
Read more