If your support agent asks you to create a support bundle for troubleshooting purposes, follow this procedure. The bundle creation process extracts information from your VAST Cluster into a compressed file that you can send to us. There is also an option to download support bundle files to your own machine.
You may choose to obfuscate sensitive data.
By default, data is collected from all CNodes and DNodes. Optionally you can restrict it to specific nodes.
The previous bundle file on each node is overwritten when you create a new one, unless you choose the option to create an aggregated bundle. Aggregated bundles contain all data from all nodes included, are saved on the management CNode and are not overwritten.
From the left navigation menu, select Support and then Bundles.
Click Create Support bundle.
In the Create Bundle dialog, complete these fields:
From the dropdown, select the best preset for the issue. Each preset defines a set of data objects to collect for a different type of issue. Verify with your support engineer which preset is suitable for your needs. By default, only management logs are collected.
Enter an identifying label to include in the bundle file name.
The prefix can contain up to 64 characters. Special characters are not allowed.
For example, if the prefix is MyUrgentIssue, a bundle using the mini preset created at exactly 7:45 am UTC on 3rd September 2020 on a cluster called cluster1 would be named bundle-mini-cluster1-MyUrgentIssue-2020-09-03-074500.tar.
Specify a start time in UTC+0 for the period from which to collect information.
Specify an end time for the period to collect information. See also Start time.
It's usually advisable to consult with the support engineer on the time frame so that you can make it as short as possible and minimize the size of the bundle.
Converts all bundled objects to text and obfuscates them. Any data that cannot be converted to text is not included in the bundle. The following types of information are replaced with a non-reversible hash: file and directory names, IP addresses, host names, user names, passwords, MAC addresses.
Saves an aggregated bundle file on the management CNode.
By default, a separate bundle file is created on each CNode and DNode that is included in the bundle. This file includes all nodes by default, or you can restrict it to specific nodes in the Advanced section (see step 4).
The aggregated file is created in addition to the separate bundle files per node. It supplies all the information in one larger file.
Unlike the separate bundle files, an aggregated bundle is not overwritten when a new support bundle is created.
Optionally, select Advanced to further specify which data to collect, and also configure upload settings:
Enable this setting to collect data from CNodes only.
Enable this setting to collect data from DNodes only.
Select specific CNodes to collect data from those CNodes and not from the other CNodes. This selection is ignored if DNodes only is selected.
Select specific DNodes to collect data from those DNodes and not from the other DNodes. This selection is ignored if CNodes only is selected.
Select VIP pools to collect data only from CNodes that are included in these VIP pools.
Enter any arguments to pass to Hubble, per instructions from the support engineer.
Enter any arguments to pass to Astron, per instructions from the support engineer.
Enter a maximum data limit to apply to the collection of binary trace files, in GB. You can use this option to limit the overall size of the support bundle.
Converts all bundled objects to a text format. Any data that cannot be converted to text is not included in the bundle.
This setting is enabled automatically if Obfuscated is selected.
Uploads the support bundle immediately.
Uploads a non-aggregated support bundle via VMS. Otherwise, the upload is done from each node.
Specifies the S3 bucket name of the destination S3 bucket.
Specifies the access key of a valid key pair to access the destination S3 bucket.
Specifies the secret key of a valid key pair to access the destination S3 bucket.
Click Create and then click Yes to confirm.
The bundle file is then created. The status of the bundle creation process is displayed on the Bundles page.
When bundle creation is completed, the bundle can be uploaded to VAST Data Support.
You can upload the support bundle to VAST Data Support immediately upon bundle creation or at a later time.
Aggregated bundles are not overwritten and therefore can be uploaded even after another bundle was created. Bundles that were not aggregated are overwritten by later bundles and therefore cannot be uploaded at a later time.
To have a support bundle uploaded immediately upon its creation, complete the upload settings in the Advanced pane in the Create Bundle dialog.
To upload a created support bundle at a later time:
In the Actions column, click the button for the bundle and select Upload.
The Upload bundle to S3 dialog appears.
The upload begins.
While the upload is in process, you'll see the Activities icon at the top of the page. Once upload is complete, it will disappear. The upload task is logged in the Activities page, which you can open from the menu.
You can download a support bundle to your local machine. Aggregated bundles are not overwritten and therefore can be downloaded even after another bundle was created. Bundles that were not aggregated are overwritten by later bundles and therefore cannot be downloaded.
In the Actions column, click the button and select Download.
The bundle file is downloaded to your machine.
When you delete a support bundle, the corresponding bundle files are deleted from the nodes, if they weren't already overwritten by a later bundle, and the bundle entry is removed from the displayed list.
Do not try to delete a support bundle before it is completely created. This will cause unexpected behavior.