Something happened with audits?

[Raptor]

The same thing is happening to me. Please fix this, Storj devs.

[therealmatrix]

happy to be not the only one who have this problem.
I want to rebuild the device in general because at the moment it was a quick fix just to get the node running.

[RecklessD]

Auditing is effected as well.

Something happened with audits?

and bloom filters.

No bloom filter from AP1 since 2025-11-11?

[bre]

Hello @therealmatrix ,

Welcome to the Forum!
I’ve passed this on to the team who will review this as well as @Raptor 's and @RecklessD 's comment of the same.

[Vadim]

i have new node 12 days old over 100Gb in size and 21 in trash, but 0 audits at all.

[BrightSilence]

Just want to add a little nuance from my observations. I’m pretty sure audits are happening as I have two new nodes and they do report audit bandwidth usage. (This also explains why @Th3Van had plenty audits in the logs.) However, it seems satellites are not updating reputation information based on those audits. Both the audit_success_count and audit_total_count in the node db’s aren’t being updated. I wasn’t aware there was a broader issue at first so I wrote some script to decode the protobuf data in audit_history as well, only to find the same result. Furthermore, I have another node that was down for 4 days for migration, yet the uptime is still at 100%.

So the nuance. It’s not that audits aren’t happening, it’s that at the least the nodes don’t get updates related to those audits, reputation, etc.

Edit: Additionally worth mentioning; I did receive notifications that my node was offline during the migration (on the 7th and 8th of December). This suggests that the satellites are still updating the reputation information on their end (at least partially), but not sending that data back to the nodes.

[RecklessD]

@BrightSilence

I’m pretty sure audits are happening as I have two new nodes and they do report audit bandwidth usage.

On this Node, there is Audit traffic. But somewhat random. Looking back at October, the figures are more consistent, and no 0’s

                  Satellite:  1wFTAg  121RTS  12EayR  12L9ZF 
Qty GET_AUDIT - 2025-12-01 :     30    2003    1073     267  
Qty GET_AUDIT - 2025-12-02 :      0       0     500      85  
Qty GET_AUDIT - 2025-12-03 :     90    1181     293      84  
Qty GET_AUDIT - 2025-12-04 :    105     984     146      95  
Qty GET_AUDIT - 2025-12-05 :      1       0       0       0  
Qty GET_AUDIT - 2025-12-06 :      0       0       0       0  
Qty GET_AUDIT - 2025-12-07 :      0       0       0       0  
Qty GET_AUDIT - 2025-12-08 :      0       0     198       0  
Qty GET_AUDIT - 2025-12-09 :    146    1630     699     310  
Qty GET_AUDIT - 2025-12-10 :     96    2995     590     380  
Qty GET_AUDIT - 2025-12-11 :    140    1076     759     171  
Qty GET_AUDIT - 2025-12-12 :      7       0     722       0  
Qty GET_AUDIT - 2025-12-13 :      0       0     742       0  
Qty GET_AUDIT - 2025-12-14 :      0       0     706       0  
Qty GET_AUDIT - 2025-12-15 :    208    1074     809     189  
Qty GET_AUDIT - 2025-12-16 :     27     478      98      56  

@BrightSilence

This suggests that the satellites are still updating the reputation information on their end (at least partially), but not sending that data back to the nodes.

I agree. But I thought the reputation updates come from the satellite, if so, they have been sending the same (not updated) information for a while now.

[Awful_Charlie]

That’s suddenly looking a lot better on my nodes

[RecklessD]

Saltlake has just jumped to 100% Online Status - on all of my nodes.

[Roberto]

Not only Saltlake, but also the other satellites, I had nodes stuck at 92% for a month and as if by magic this morning I find almost everything 100% online

[Vadim]

looks like audits started to be reported, as i got today all audits reported to node at once.

[HGPlays]

I see that it also have updated audit scores, but i have a weird thing:

I see on eu: Vetting result pending

Meaning it has all audits but is still only geting 1% traffic - seems off?

[yuz98]

My saltlake node dropped from 100% to 0% in 1 day.

[BrightSilence]

I can confirm my node has many audits now as well. Seems something or someone reminded the satellite to update our nodes. I didn’t have much time to get screenshots this morning before leaving for work. But yeah, seems to work again.

[Alexey]

Now it’s not bound to 100 audits, it’s dynamic. The goal is the same - vetting should be not less than a month.
See

What is the new vetting time?

The main goal of vetting system is to prove the stability of the new nodes for a longer period of time. Required number of audits are bumped as we have more audits than earlier. To avoid too quick vetting, we also bumped the required number of audits with an estimation to be vetted in 3-4 weeks. This will be adjusted more precisely with this patch (can require either a number of audit or a minimum period of time): 94c347e64 N - satellite/reputation: check node age in addition to number of aud…

[Toyoo]

Ok, looking at scores of my flock of nodes, the problem seems to have been remedied. That said, judging from the what my nodes report now, I understand we should be seeing:

• Large singular jumps in online score—online score was correctly tracked by a satellite, just not sent to nodes. So what you see now is effectively accummulated score over the last month.
• No jumps in audit scores—audits over the last month may have been not actually tracked, or if tracked, the past month was intentionally ignored. If your node was about to disqualify, consider this an early Christmas present

I was starting to get worried about the state of my nodes, glad to see this corrected!

[yuz98]

Online Score is weird, screenshotted same day. Saltlake drop from 100 to 0 then now is 51. others drop while my node was online. The audit score dropped 0.04% as well on ap1.

[Alexey]

@yuz98

The audit score dropped 0.04% as well on ap1.

This may indicate data loss.

[RecklessD]

@yuz98

The audit score dropped 0.04% as well on ap1.

Nothing to worry about.at 0.04% (It’s an issue if it gets down to 96%) it will be back up in a couple of days (provided Audits continue to work)

@yuz98

Saltlake drop from 100 to 0 then now is 51

Saltlake does strange things sometimes.

[BrightSilence]

Wasn’t aware of that. Thanks for letting us know. Working on an update for the earnings calculator to incorporate this so nodes don’t say “Vetting results pending” for weeks, but instead display the lowest value between vetting progress based on counts and vetting progress based on time. I asked in the topic you linked whether the release defaults of 100 audits and 504 hours are being used, but since they usually are I might release the new version before confirmation. (Can always fix it later)

@HGPlays don’t worry about seeing vetting result pending in the current version. I have to derive it based on what the node knows and it isn’t aware of the exact conditions required to be vetted, so I have to manually add those conditions in the calculator. New version will show a percentage that reflects both conditions.