Unsolved
24 Posts
4
8229
list of bugs/issues with OM Enterprise
I just wanted to raise the following issues, I would be interested to know if they could be patched in later versions:
- When a firmware baseline is created against Dell firmware repository, if there is a group of nodes in the baseline, you have to individually go through each device to select the firmware that you want to update. For example if you have 3 available updates for all 100 devices in the baseline (all the same model) and I only want to run the iDrac update for all of them, I have to individually go to each device and untick the other updates... This is unmanageable and the only current way around this I can see is to use DRM (which I don't think should be a requirement).
- When I do a firmware baseline for a device - say a C6420 - the updates will be downloaded in a package when I run the firmware upgrade. If I then want to do the same update on more C6420's in the future, the firmware will be re-downloaded each time. This is a huge drain on time since the update has already been downloaded and it seems to me that when updates are gathered they should be stored locally so that they can be used on other devices without fetching each time. This is not helped by Dells repository servers being very slow, but would be less of an issue if when an update has been downloaded once, it didn't need to be done again.
- There is no way to have multiple IPs for an OME instance. For example I may want to put the OME instance on two networks to give it access to two layer 2 networks - this is not currently possible (I'm using the Linux version)
- under device lists such as 'all devices', or manually created groups there is not a feature to select all pages, which other parts of OME do have. there is only the option to select all devices on the current page, which is only 25 devices. I may want to do a search for all devices with the ip starting 10.4.3 for example and then the 200 devices under there add them all to a custom group and currently I would have to add them in pages of 25 which takes a long time. There are other sections in OME where multiple pages of objects can be selected so I think this one would be a quick fix?
- there is no feature for adding custom scripts for alerting. For example I want to add a slack web hook so that I can get critical alerts to a slack channel. This is far more useful than email alerts so its a shame that this isn't currently possible.
- The hardware logs are not listed for 13G devices such as C6320's or C4130's - surely this is possible in the same way that it is with 14G devices such as C6420, as this is just listing a log list from the device.
- in advanced filters there is no option to query a closed group. for example in the all devices group i may want to search for only hosts 1 to 10, but there is no search filtering such as in the hostname field writing host-[1-10] where the hostnames for the devices are 'host-1, host-2'.
- There are many features missing from devices in comparison to logging on to the idrac of the device. Could the options on a device not be similar to that of logging onto individual web consoles of devices? This would make managing individual devices much easier as currently there is still requirement to log on to individual web consoles.
I do find the software very useful, but there are certainly some ironing out and features to add such as the above to really make it a great tool - which I think it could be.
ElectricalTears
15 Posts
2
October 31st, 2018 17:00
Some others that I've come across.
vSwizz
24 Posts
0
November 1st, 2018 02:00
Good points. I also just remembered:
vSwizz
24 Posts
0
November 1st, 2018 14:00
Just tonight I have discovered another big issue:
DELL-Pupul M
1K Posts
0
November 1st, 2018 23:00
Thanks guys for the post. Lot of what you have mentioned is how the product was designed and we are working towards making it better as and when we get more requests from customers. Please have these requests coming. I will pass this on to the right guys to see what can be done about making the product in future release.
vSwizz
24 Posts
0
November 8th, 2018 13:00
An update on the firmware catalogue refresh for OME:
From talking to Dell it appears that the catalogue is updated 'once or twice per month'.
I've just done a re-scan of the firmware catalogue and it updated on the 30th of October (the previous one was from the 15th), some questions on this:
When you are putting firmware on 2000 nodes you don't want to create a baseline on the 21st only to find out a new update came out for the nodes 6 days ago but hasn't been updated on the OME catalogue. You then have to do another upgrade on all of the nodes which was already out when you did the first upgrade. This means that we will have to continue to rely on checking the website for what updates are out, which seems counter intuitive as searching individual service tags on the Dell website is manual and time consuming in comparison to checking OME (if it was up-to-date consistently).
At the very least I think that there needs to be a clear advertisement of when the catalogue is updated (such as once per day/week/month), if we will have to manually delete the old catalogue and search for a new one - though automatic and updating very regularly is still my preference!
B1G_WH1T3
10 Posts
0
November 27th, 2018 02:00
Hi
Some issue I also picked up, our estate is compiled out of Rx30 & Rx40, we also have some older hardware Rx10,20, which I've seen does not work too nice with OME which is understandable.
Issues :
Alerting :
Subject/alert needs to be more descriptive - example we had a M630 with a uncorrectable memory issue, this caused the host to reboot, OME generated an warning alert with CDEV6129 device health has deteriorated.
Technically this should been alerted as a critical hardware alert as one of the dimms is faulty and caused a system reboot.
An alerting glossary would be nice, like which $variables we can use in the template to customize our alerting body's and or subject lines.
Warranty :
Warranty issues have been listed here already also getting the issue where it reloads the main page and does not actually take me to the warranty screen.
one big issue I have is the warranty is showing as expired, where as we already renewed it and online it shows as renewed, the bug I found is that OME does not remove the expired warranty and thus it looks like the hardware has expired warranty's.
Compliance :
When I export a device status and customize the XML to look for example NTP settings, and I point it to a static group consisting of mixed hardware, it only displays the latest hardware in that group, even though the export was done on older hardware.
Hardware names issue also detected, I have to change it from Prefer DNS to Netbios, then back to Prefer DNS, to fix names display in list.
Getting random SMB issues on hardware that worked without SMB V1, some export works fine, other times the same hardware gives issue and I have to enable SMB v1 to get the report done and then disable it again.
Firmware :
Would it be able to integrate software updates also, like what the SUU iso does, updating firmware and drivers of a physical server running and operating system like server 2016 etc
Not sure if this belongs with this product, but some random requests I have
Using dashboard as a monitoring tool
- Enable custom widgets
- Enable timeout period/ do not auto logoff
- issue i have is to clear alerts off dashboard, alerts are marked as acknowledged but they still stay on the screen
multiple instance of logins, example if you middle click on a device it opens a new tab and you have to login again instead of saving the current session.
Reporting
Please enable scheduled reports on the default ones found and or enable it for custom reports also.
Reporting issue picked up :
Simple report consist of device name, device model, service tag, power, server total memory and processor model, with these simple selections it causes to double the entries, some servers repeat up to 4 times, I have examples of them repeating up to 7 times with no explanation to why they are doing it.
fa_jking
1 Message
0
November 30th, 2018 07:00
A blocker for me is that if the CMC password contains an ampersand, the firmware update fails due to (what appears to be) not properly escaping the password when the ssh/telnet session is executed.
geruta
11 Posts
0
December 2nd, 2018 06:00
I have issues and feedback that I will write out in two different sections.
Feedback/Suggestions:
---After issues are corrected in the iDRAC (be it DIMM replacement, HD, etc.) why don't the alerts automatically clear? It seems counter intuitive to see an alert in OME that has already been addressed and no longer is "flagged" in the system's iDRAC.
---Instead of asking users for a list of IPs in the script execution tool, allow users to select from a list of devices or groups. Additionally, I'd love to see the ability to pre-populate a set of scripts in the various "banks" and then right click on a group or iDRAC from the all devices menu and select "run script [1, 2, 3...]."
---Why does OME need so many ports to be fully functional? It is not possible to push firmware, receive device health alerts, any many other functions from the 443 port? Its cumbersome to open rules with the firewall teams for ports 111, 137, 138, 139, 443, 445, 162, 892, and 2049 to get the ability to push firmware, push iDRAC/BIOS configs, receive SNMP traps, and manage the device remotely. Just having one port would go a long way to simplify administration and likely leading to better adoption. Just have a quick look on this forum and several users are complaining about connectivity issues due to the number of ports needed.
---Why do I need to install a separate appliance for call home (Support Assist)? This is something that can and should be integrated into OME.
---Why do I need to install a separate appliance for managing repositories (Repo Manager)? This is something that can and should be integrated into OME.
Issues:
---iDRAC systems are constantly showing the alert "CDEV6131 - Connection to the device was not successful." These are coming in at a rate of about one alert every five minutes. Adjusting the device health poll from 1 hour to 20 mins, to 2 hours doesn't change the rate at which they flow in.
---AD Auth breaks after upgrading or installing a fresh 3.0 appliance as you can no longer enter the insecure AD port. I don't own AD in my environment and I can't force them to enable the insecure port.
---When clicking on a server in OME and getting directed to the device's page, it takes a solid 1-2 mins to load the device health. This should be instant or no longer than 5-10 seconds.
MarcLaflamme
15 Posts
0
December 3rd, 2018 12:00
We're getting this after the upgrade. Thinking I should just blow it away and reinstall...
acpp
22 Posts
0
April 25th, 2019 11:00
Same happened to me. I have to do a fresh install...