<div class="socmaildefaultfont" dir="ltr" style="font-family:Arial, Helvetica, sans-serif;font-size:10.5pt" ><div dir="ltr" >Hello Sylvain</div> <div dir="ltr" > </div> <div dir="ltr" >thanks for the reply.</div> <div dir="ltr" > </div> <div dir="ltr" >The devices in question are Cisco and Brocade-branded Fibre-channel switches. For Cisco the primary interface is SNMP, for Brocade, it is currently SMI-S to the Brocade software management tool called Network Advisor. For Cisco, the user would have to input the connection information for at least one switch, the rest of the fabric could be discovered based on data from that switch. For Brocade, you would be querying a software database.</div> <div dir="ltr" > </div> <div dir="ltr" >We do have some existing tools that can run commands remotely on a switch and collect the text output, but that output then has to be parsed.</div> <div dir="ltr" > </div> <div dir="ltr" >As far as using the API, I was hoping it could be used to input an already created topology or other parsed data to Skydive and have Skydive draw it. For the topology request here:</div> <div dir="ltr" > </div> <div dir="ltr" ><a href="http://skydive.network/documentation/api#topology-requests">http://skydive.network/documentation/api#topology-requests</a></div> <div dir="ltr" > </div> <div dir="ltr" >does that request that Skydive build a topology from probed data or does it tell Skydive what the topology is and Skydive produces a diagram of it? One of the projects I'm interested in using Skydive for, the data has already been collected from customer networks, we would need a way via JSON or similar to get it into Skydive. For this use-case there isn't an opportunity to probe the network.</div> <div dir="ltr" > </div> <div dir="ltr" >Some of the fields could be used as-is, but others such as 'Type' may need enhancements. As an example, I didn't see a type of 'Switch' listed as a valid Type.</div> <div dir="ltr" > </div> <div dir="ltr" ><div class="socmaildefaultfont" dir="ltr" style="font-family:Arial, Helvetica, sans-serif;font-size:10.5pt" ><div dir="ltr" ><font size="3" >thanks,</font><br><br><b><font color="#7C7C5F" face="Arial" size="3" >David M. Green</font></b><br><b><font color="#7C7C5F" size="3" >Advisory Software Engineer</font></b><br><b><font face="Arial" size="2" >SAN Troubleshooting and Performance, IBM Solution Central Support</font></b><br><b><font color="#336699" face="Arial" size="1" >Phone: </font></b><font face="Arial" size="1" >1-919-486-2185</font><br><b><font color="#336699" face="Arial" size="1" >E-mail: </font></b><a href="mailto:greendm@us.ibm.com"><u><font color="#555555" face="Arial" size="1" >greendm@us.ibm.com</font></u></a></div> <div dir="ltr" > </div> <div dir="ltr" ><a href="https://www.ibm.com/developerworks/community/blogs/e33368ad-90f1-4e20-a38f-2f93f38d39c5?lang=en"><img alt="" src="https://www.ibm.com/developerworks/community/blogs/e33368ad-90f1-4e20-a38f-2f93f38d39c5/resource/followblogtransparentrz.png" ></a><br><br><img height="58" src="cid:15369316241721" width="258" ><br> </div></div></div> <div dir="ltr" > </div> <div dir="ltr" > </div> <blockquote data-history-content-modified="1" dir="ltr" style="border-left:solid #aaaaaa 2px; margin-left:5px; padding-left:5px; direction:ltr; margin-right:0px" >----- Original message -----<br>From: Sylvain Afchain <safchain@redhat.com><br>To: greendm@us.ibm.com<br>Cc: skydive-dev@redhat.com, Eran Raichstein <ERANRA@il.ibm.com><br>Subject: Re: [Skydive-dev] Skydive Support of Fibre-Channel Networks<br>Date: Fri, Sep 14, 2018 3:52 AM<br> <div dir="ltr" ><div dir="ltr" ><div dir="ltr" >Hello David, <div> </div> <div>You're right currently Skydive only report what is on the host but physical equipment discovery is something we are going to work on. What kind of protocols you plan to use for the discovery ? it will help us to provide them within Skydive.</div> <div>That said, you can use the Skydive API, there is a python library[1] that can be used. We recently developed an Ansible module[2] to add</div> <div>node/edge to the Skydive topology (a blog post/doc is coming). We used it to report LLDP information. There is also a new set of command line[3] that can be used but as it has been added recently there is no documentation yet.</div> <div> </div> <div>For the WebUI and the edges, you can add multiple edges between nodes but I think that in the WebUI only one will be visible (overlapping). That is something we can fix. For the "on_click" thing, you will be able to click on the edge to get the metadata of the edge.</div> <div> </div> <div>We are currently working on a new version of the WebUI, any suggestion/idea will be appreciate :)</div> <div> </div> <div>Thanks,</div> <div>Sylvain</div> <div> </div> <div>[1] <a href="http://skydive.network/documentation/api-python" target="_blank">http://skydive.network/documentation/api-python</a></div> <div>[2] <a href="https://github.com/skydive-project/skydive/tree/master/contrib/ansible/library" target="_blank">https://github.com/skydive-project/skydive/tree/master/contrib/ansible/library</a></div> <div>[3] skydive client node-rule, skydive client edge-rule</div></div></div></div> <div><div dir="ltr" >On Fri, Sep 14, 2018 at 9:28 AM David Green <<a href="mailto:greendm@us.ibm.com" target="_blank">greendm@us.ibm.com</a>> wrote:</div> <blockquote style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" ><div dir="ltr" style="font-family:Arial,Helvetica,sans-serif;font-size:10.5pt" ><div dir="ltr" >I'm evaluating Skydive as a possible tool to use for a project I am working on with fibre-channel SANs. I know Skydive currently only works with ethernet networks. It also gets most of its information from the hosts on the network whereas with fibre-channel most of the information comes from switches. Is it possible enhance Skydive to work with fibre-channel? My primary interest is using Skydive as a graphing tool and being able to draw all or part of the network depending on what the user wants to examine. Can the Skydive analyzer component be used to draw a set of nodes and edges when they are passed in via the API, or does the analyzer take nodes and still do analysis on them to determine where the edges are? </div> <div dir="ltr" > </div> <div dir="ltr" >Is there flexibility to be able to label each end of an edge and to be able to have an edge respond to on_click events? For instance, there may be several connections between two nodes, and a user may want to mouse over an edge, or click it to expand it out into its component links. </div> <div dir="ltr" > </div> <div dir="ltr" > </div> <div dir="ltr" ><div dir="ltr" style="font-family:Arial,Helvetica,sans-serif;font-size:10.5pt" ><div dir="ltr" ><font size="3" >thanks,</font><br><br><b><font color="#7C7C5F" face="Arial" size="3" >David M. Green</font></b><br><b><font color="#7C7C5F" size="3" >Advisory Software Engineer</font></b><br><b><font face="Arial" size="2" >SAN Troubleshooting and Performance, IBM Solution Central Support</font></b><br><b><font color="#336699" face="Arial" size="1" >Phone: </font></b><font face="Arial" size="1" >1-919-486-2185</font><br><b><font color="#336699" face="Arial" size="1" >E-mail: </font></b><a href="mailto:greendm@us.ibm.com" target="_blank"><u><font color="#555555" face="Arial" size="1" >greendm@us.ibm.com</font></u></a></div> <div dir="ltr" > </div> <div dir="ltr" ><a href="https://www.ibm.com/developerworks/community/blogs/e33368ad-90f1-4e20-a38f-2f93f38d39c5?lang=en" tabindex="-1" target="_blank"><img alt="" s_org_src="https://www.ibm.com/developerworks/community/blogs/e33368ad-90f1-4e20-a38f-2f93f38d39c5/resource/followblogtransparentrz.png" ></a><br><br><br> </div></div></div></div><br>_______________________________________________<br>Skydive-dev mailing list<br><a href="mailto:Skydive-dev@redhat.com" target="_blank">Skydive-dev@redhat.com</a><br><a href="https://www.redhat.com/mailman/listinfo/skydive-dev" rel="noreferrer" target="_blank">https://www.redhat.com/mailman/listinfo/skydive-dev</a></blockquote></div> <div id="MIMEAttachInfoDiv" style="display:none" title="jpeg|Image.15368708282500.jpg" > </div></blockquote> <div dir="ltr" > </div></div> <BR>