[Working Blog] ONTAP Snapshot Reporting in Python - Part 8 - Creating the Report

 One addition to part 5 is to get the snapshot create_time. We can do that this way:

Snapshot.get_collection(volUUID,fields="create_time")

Note: I need to think about how we use "create_time" since currently I'm just recording a list of snapshots by name.

Report Fields

The original report idea (back in part 1) had these fields:

1) Snapshot Name, 2) Snapshot ID (if any), 3) Backup Schedule, 4) Volume Name/Path, 5) Backup Results, 6) Time, 7) Success/Failed

For the report version 1, I'm going to change this to:

1) SVM, 2) Volume, 3) Snapshot name, 4) Snapshot policy, 5) Success (exists), 6) Create Time

We have all the information above, now to form it into a report.

Getting the Output

The process:

• for each volume:
•  find snapshot policy
• loop through snapPolList
• if we have the snapshot, record in the report as success
• if not, record it as failure

This should work:

import csv

outputCSV = []
outputFileName = 'snapshotReport.csv'

for u in volUUIDs:
  if volByUUID[u]['style'] != 'flexvol':
    # Only considering FlexVols (the idea is to skip FlexGroups)
    continue
  sp = volByUUID[u]['snapshot_policy']
  spUUID = snapPolicyNameToUUID[sp]
  expectedSnaps = snapPolSnapList[spUUID]
  for s in expectedSnaps:
    if s in set(snapsByVolUUID[u]):
      success = "SUCCESS"
      createTime = str(snapDataByVolUUID[u][s]['create_time'])
    else:
      success = "FAILURE"
      createTime = ""
    outDict = {}
    outDict['SVM']              = volByUUID[u]['svm']
    outDict['Volume']           = volUUIDtoName[u]
    outDict['Snapshot Name']    = s
    outDict['Snapshot Policy']  = volByUUID[u]['snapshot_policy']
    outDict['Success (exists)'] = success
    outDict['Create Time']      = createTime
    outputCSV += [outDict]
    

with open(outputFileName, 'w', newline='') as csvfile:
  fieldnames = ['SVM','Volume','Snapshot Name','Snapshot Policy','Success (exists)','Create Time']
  writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
  writer.writeheader()
  for r in outputCSV:
    writer.writerow(r)

I think that is good (need to test.)

Getting Snapshot 'Create Time' Working

Now, how about getting the 'Create Time' working?

## 7) ACQUIRE SNAPSHOTS ##

snapsByVolUUID = {}
snapDataByVolUUID = {}

for u in volUUIDs:
  snaps = Snapshot.get_collection(u,fields="create_time")
  snapsByVolUUID[u] = []
  snapDataByVolUUID[u] = {}
  for s in snaps:
    snapsByVolUUID[u] += [s.name]
    snapDataByVolUUID[u][s.name] = {}
    snapDataByVolUUID[u][s.name]['create_time'] = s.create_time

Which allows us to amend the above (getting the output) with 'Create Time' 

= str(snapDataByVolUUID[u][s]['create_time'])

Problem

I notice the snapshot naming convention is using the cluster timezone, not UTC which is what I had thought (since my cluster is set to Tasmania timezone and I'm 10 hours away, it is very noticeable). I'll have to fix this...

Image: ONTAP Snapshots Named By Cluster Timezone

Fixing the Time Issue

pip install pytz

import pytz

clus = Cluster()
clus.get()

utcDateTime = clus.statistics.timestamp
local_tz    = pytz.timezone(clus.timezone.name)
localDT     = utcDateTime.astimezone(local_tz)

With the above, all we'll need to do is change the usage of utcDateTime to localDT.

Next Steps

In Part 9 we'll bring it all together:

• Code from part 5.
• Code from part 7.
• Code from part 8 (including section 7 fixes and time fixes)
• And turn it into a script.

That might be too much for one part.

To be continued ...