Recent shard lag
The 2-second-lag-spikes-every-90-seconds is an issue we are working to resolve. Unfortunately, to properly address such issues, it is occasionally necessary to collect live performance samples from Hybrid.
These recent performance samples have caused a few longer pauses over the past few days. You may have noticed that the pauses have generally occurred during or after a world save--this is because we manually invoke a save to ensure no data is lost in case the sample fails or must be aborted.
Fortunately, however, we've collected enough data to identify what has been causing the initial lag. The Hybrid core has been updated with a number changes that will help reduce the problem, as well as with means to collect better performance data without the huge delays.
You can expect the shard to run smoother going forward. And while there may still be periodic issues, keep in mind that we will continue to monitor and tune performance as necessary.
Thanks,
The 2-second-lag-spikes-every-90-seconds is an issue we are working to resolve. Unfortunately, to properly address such issues, it is occasionally necessary to collect live performance samples from Hybrid.
These recent performance samples have caused a few longer pauses over the past few days. You may have noticed that the pauses have generally occurred during or after a world save--this is because we manually invoke a save to ensure no data is lost in case the sample fails or must be aborted.
Fortunately, however, we've collected enough data to identify what has been causing the initial lag. The Hybrid core has been updated with a number changes that will help reduce the problem, as well as with means to collect better performance data without the huge delays.
You can expect the shard to run smoother going forward. And while there may still be periodic issues, keep in mind that we will continue to monitor and tune performance as necessary.
Thanks,