Change Details

@Addshore alerted us that from [[ https://grafana.wikimedia.org/d/000000574/t204083-investigation?orgId=1&from=now-14d&to=now&panelId=3&fullscreen&edit | his dashboard ]] it seems that tx bandwidth usage for mc1033 spiked on 2019-05-02 ~19:40 UTC. In T97368 a similar pattern was observed and investigated, but it might not be related. From https://tools.wmflabs.org/sal/log/AWp6A7RAEHTBTPG-zSOo it looks like 1.34.0-wmf.3 was deployed around that time, so new keys might be involved.

@Addshore alerted us that from [[ https://grafana.wikimedia.org/d/000000574/t204083-investigation?orgId=1&from=now-14d&to=now&panelId=3&fullscreen&edit | his dashboard ]] it seems that tx bandwidth usage for mc1033 spiked on 2019-05-02 19:26 UTC. {F29166140} In T97368 a similar pattern was observed and investigated, but it might not be related. From [SAL](https://tools.wmflabs.org/sal/production?p=0&q=&d=2019-05-02), we see 1.34.0-wmf.3 was deployed at that time, so new keys might be involved.

@Addshore alerted us that from [[ https://grafana.wikimedia.org/d/000000574/t204083-investigation?orgId=1&from=now-14d&to=now&panelId=3&fullscreen&edit | his dashboard ]] it seems that tx bandwidth usage for mc1033 spiked on 2019-05-02 ~19:4019:26 UTC. {F29166140} In T97368 a similar pattern was observed and investigated, but it might not be related. From [SAL](https://tools.wmflabs.org/sal/log/AWp6A7RAEHTBTPG-zSOo it looks like 1production?p=0&q=&d=2019-05-02), we see 1.34.0-wmf.3 was deployed aroundt that time, so new keys might be involved.