Jump to content

ttl

Members
  • Posts

    23
  • Joined

  • Last visited

Recent Profile Visitors

102 profile views

ttl's Achievements

Apprentice

Apprentice (3/14)

  • One Year In
  • One Month Later
  • Collaborator Rare
  • Week One Done
  • First Post

Recent Badges

0

Reputation

  1. Thanks for the tip, is there a way to query for unused filesets, or those associated to a small volume of machines? My question was more along the lines of, I don't know if there's any legacy data like old models and such that could be purged away. I never explored the innards of Filewave because I'm not all that interested in breaking it
  2. We recently did the upgrade to Debian and found that Datto currently doesn't support Debian 12. Planning to utilize the backup script for the time being. Our Filewave setup is rather large, is there any way to reduce it's size? The goal is to make these backups much less resource intensive, if it's possible.
  3. It's a python script running on a rendezvous server used for small jobs with APIs. device_id is obtained from: filewave.example.com/api/inv/api/v1/query_result/123 This ended up being a programming bug on my part, something like this was coming from Snipe It: value_three = snipeit['expected_checkin'] But Snipe It was sending it in json format: {'date': '2024-05-02', 'formatted': '2024-05-02'} So value_three should have been: value_three = snipeit['expected_checkin']['date'] Devices that previously triggered a 500 now appear to be updating successfully. (The query numbers are different on each refresh, which is expected with the change in data)
  4. Assigned to all devices. I started logging these and having a report email me everytime these are triggered, seems to be the same devices everytime. No similarity between them I can ascertain. I have a script that runs on a cron every 30 minutes that hooks into this system and bring data back to Filewave, I'll adjust the time slightly in case it is hitting the smart group recalculation, thanks!
  5. This doesn't happen with every device, but only some of them. Is there anywhere to look on Filewave's side to see what would cause the custom field update to issue a 500 response? Here's the component that works for virtually every device in Filewave, short of a small handful: # update the custom field using the device id api = filewave + 'inv/api/v1/client/' + device_id params = { "CustomFields": { "customfield_one": { "exitCode": 0, "status": 0, "updateTime": datetime.now().isoformat(), "value": value_one }, "customfield_two": { "exitCode": 0, "status": 0, "updateTime": datetime.now().isoformat(), "value": value_two }, "customfield_three": { "exitCode": 0, "status": 0, "updateTime": datetime.now().isoformat(), "value": value_three } } } params = json.dumps(params) update_fields = requests.patch(api, params, headers=headers, timeout=2)
  6. Perfect, I can see everything I need in Postman so I should be able to script with this. Will there be a public API in the future, or is this a niche thing no other Filewave customers use?
  7. /api/devices/v1/devices/[groupnumber] seems to only return the group details. Is there a way to get all of the devices under groupnumber?
  8. ttl

    Date Format

    That ended up being the problem! My test Mac is also the same Mac I do app signing on, so it had Xcode (and dev tools w/ Python) and all of the machines that worked had brew or Xcode on them, too. Every machine that's custom field script was failing didn't have either. Deploying Python did the trick and the Python scripts are fully functional, now!
  9. ttl

    Date Format

    Wow it was literally right in front of my face! I have a python script that when I run it manually on the machine it works, but when it's sent as a custom field script I get failure on about 90% of the machines and I can't figure out why. One would think it'd fail for all or none. No common denominator between the successes and failures (both success & failures are Sonoma, for example) I'd originally erred it being the date format but that is most definitely not the cause. What would be the best approach to troubleshooting such a thing? The client log seems to only have an error code.
  10. ttl

    Date Format

    It ended up taking 2022-04-13 08:04:26 Didn't realize I left a debug print() in my code. Where are custom field scripts kept on the client machine?
  11. ttl

    Date Format

    When a custom field's data type is set to Date/Time, what format is it expecting? 2022-04-13 08:04:26 2022-04-13 8:04AM Or something else? Didn't find anything mentioned in the knowledgebase.
  12. Powershell scripts are executed as SYSTEM and trying to execute a PS command as a user I run into the Powershell restrictions, which I don't which to change. There's some third party options and PS modules which I had minimal success with, is there perhaps a way to utilize Filewave to execute a GUI notification? This was easily achievable in Mac using osascript and invoking su -l $current_user -c /scriptpath Windows is proving substantially more difficult to achieve the same. Use case: There's an upgrade fileset for an app heavily used by users. During the upgrade there IS a brief interruption and it'd be nice to notify them on-screen, as opposed to manually sending out emails which they often don't read until after the interruption already occurred.
  13. When testing, I like to have logs emailed to me. When users are remote, I can't use Filewave's "Get Log" function, even with the firewall ports script active (that's offered in Filewave's documentation) I have a debug variable set, but I have to go into the script and change that everytime I want to switch from debug to live mode and that doesn't target the single user I want to zero in on for logs, but anyone that's got the association.
  14. I don't seem to have an edit button for my post. While I'm not clear on where Filewave is reading the version & vendor data from, I can omit the blanks like so. ls -r "C:\Program Files\Mozilla Firefox" | sls 'Vendor=' -ca Returns: Targeting this is good enough, for now.
  15. Suspect this may be a false-positive. I got the custom field lookup working, but I'm still seeing blank versions getting pulled in: Interestingly, both "Firefox" applications are in the same install path. (Is there a way to get the filename in this view, or am I overlooking something in the Application options?) Collecting an ls from a target machine doesn't show anything peculiar and just the typical directory listing. Doing a lookup for vendor matches: ls -r "C:\Program Files\Mozilla Firefox" | sls vendor | select -u Path Gives the following result: Path ---- C:\Program Files\Mozilla Firefox\application.ini C:\Program Files\Mozilla Firefox\crashreporter.exe C:\Program Files\Mozilla Firefox\crashreporter.ini C:\Program Files\Mozilla Firefox\minidump-analyzer.exe C:\Program Files\Mozilla Firefox\mozwer.dll C:\Program Files\Mozilla Firefox\nss3.dll C:\Program Files\Mozilla Firefox\omni.ja C:\Program Files\Mozilla Firefox\softokn3.dll C:\Program Files\Mozilla Firefox\xul.dll C:\Program Files\Mozilla Firefox\browser\omni.ja C:\Program Files\Mozilla Firefox\browser\features\screenshots@mozilla.org.xpi C:\Program Files\Mozilla Firefox\browser\features\webcompat@mozilla.org.xpi
×
×
  • Create New...