<div dir="ltr"><div><div><div><div><div><div>After the crash of today, my cluster is still syncing data.<br></div>From atop command I can see that network and disks I/O activity is not so intensive:<br><br>eth0 is generating 150Mbit/s pf traffic.<br>
</div>But it's a gigabit , so it could go much faster.<br></div>Disk is busy 55%.<br><br></div>What do you think about?<br></div>Is sheepdog limiting the sync speed to leave I/O for guests?<br></div>Does sync have any relation with cache?<br>
<br><div><div><div><div><div><div><div><div><div><br>NET | eth0 15% | | pcki 124639 | pcko 25187 | | si 151 Mbps | so 1414 Kbps | | coll 0 | | mlti 60 | erri 0 | | erro 0 | drpi 0 | | drpo 0 |<br>
NET | eth1 0% | | pcki 26 | pcko 0 | | si 3 Kbps | so 0 Kbps | | coll 0 | | mlti 0 | erri 0 | | erro 0 | drpi 0 | | drpo 0 |<br>
NET | br0 ---- | | pcki 25969 | pcko 25345 | | si 146 Mbps | so 1342 Kbps | | coll 0 | | mlti 0 | erri 0 | | erro 0 | drpi 0 | | drpo 0 <br>
<br></div></div></div></div></div></div></div></div></div></div>