Massive difference between audit counts

Node Operators
1

I’m seeing massive differences between the audit counts on my node. I’d expect some of course but more than a factor of 10 seems ridiculous.

root@vm1-storj-x64:/usr/local/scripts# for sat in docker exec -i storagenode3 wget -qO - localhost:14002/api/sno | jq .satellites[].id -r; do docker exec -i storagenode3 wget -qO - localhost:14002/api/sno/satellite/$sat | jq .id,.audit; done
“12tRQrMTWUWwzwGh18i7Fqs67kmdhH9t6aToeiwbo5mfS2rUmo”
{
“totalCount”: 167,
“successCount”: 167,
“alpha”: 19.99638,
“beta”: 0,
“unknownAlpha”: 19.99638,
“unknownBeta”: 0,
“score”: 1,
“unknownScore”: 1
}
“1wFTAgs9DP5RSnCqKV1eLf6N9wtk4EAtmN5DpSxcs8EjT69tGE”
{
“totalCount”: 17,
“successCount”: 16,
“alpha”: 11.637592,
“beta”: 0,
“unknownAlpha”: 11.637592,
“unknownBeta”: 0,
“score”: 1,
“unknownScore”: 1
}
“121RTSDpyNZVcEU84Ticf2L1ntiuUimbWgfATz21tuvgk3vzoA6”
{
“totalCount”: 9,
“successCount”: 9,
“alpha”: 8.02526,
“beta”: 0,
“unknownAlpha”: 8.02526,
“unknownBeta”: 0,
“score”: 1,
“unknownScore”: 1
}
“12EayRS2V1kEsWESU9QMRseFhdxYxKicsiFmxrsLZHeLUtdps3S”
{
“totalCount”: 5,
“successCount”: 5,
“alpha”: 5.298162,
“beta”: 0,
“unknownAlpha”: 5.298162,
“unknownBeta”: 0,
“score”: 1,
“unknownScore”: 1
}
“12L9ZFwhzVpuEKMUNUqkaTLGzwY9G24tbiigLiXpmZWKwmcNDDs”
{
“totalCount”: 9,
“successCount”: 9,
“alpha”: 8.02526,
“beta”: 0,
“unknownAlpha”: 8.02526,
“unknownBeta”: 0,
“score”: 1,
“unknownScore”: 1
}
“12rfG3sh9NCWiX3ivPjq2HtdLmbqCrvHVEzJubnzFzosMuawymB”
{
“totalCount”: 4,
“successCount”: 4,
“alpha”: 4.524381,
“beta”: 0,
“unknownAlpha”: 4.524381,
“unknownBeta”: 0,
“score”: 1,
“unknownScore”: 1

2

Why do you think it is ridiculous ?

3

Because at that rate it would take more than 4 months to get vetted by the other satellites.

4

I am wondering if you are aware that audits are per satellite and are completely random. Each satellite needs 100 audits which take ~ 1 month to classify a node to be fully vetted.

5

I’m aware but the greater the time period the more those numbers will average out. Will I be able to predict the audit rate at any period of time? No. But, for example if you have a 60 day average it will give something close to a daily rate. Again, it won’t be exact. But your max and min numbers should approximate. I’m thinking you wont see a number like 100k audits in a day from one satellite so storj must have specified a range.
Currently, as it stands if some of the satellite continue even double or triple their current audit rate the satellites will not complete the required 100 audits in a month. Or two. Or three.

6

The audit rate grows with the amount of data you hold from that particular satellite.

Is that an issue? You are already vetted for one satellite which gives you full traffic. The others don’t have so many audits yet, which means there is also not so much traffic there. Even if you were fully vetted with them there would not be so much traffic compared to that first satellite.

2 Likes
7

It’s currently taking way more than a month to get vetted on europe-north.

2 Likes
8

Maybe, but ingress is also really low from this satellite. I had 70 MB yesterday.

1 Like