Search results

You will never ever connect SAS cables from NetApp to hosts. You will create Aggregates and Volumes and maybe LUNs on the NetApp Controllers/Cluster. Is this a 7mode oder cdot ONTAP? The best would be export with NFS und mount the NFS share on both nodes. Then you have snapshots depening on...

Ah i think you need to check the bios settings there is an option i had to set with i think x520 maybe the same for x710 I think this was map memory above 4g

Should work with ulimit. I think the apache config is wrong or having real heavy load. Maybe an attack on the apache?

Arista 7150s a real nice ulta low latency switch. edit: as i see i have an older fw version but this should not be a problem: root@pmxn1:~# ethtool -i eth3 driver: i40e version: 1.4.25-k firmware-version: 4.53 0x80001dc0 0.0.0 bus-info: 0000:40:00.0 supports-statistics: yes supports-test: yes...

Try to update the FW of the card. Do you have Patchpanels between switch and server? If yes try to cross the cable maybe rx and tx are not correct. https://downloadcenter.intel.com/de/download/24769/NVM-Update-Utility-for-Intel-Ethernet-Converged-Network-Adapter-XL710-X710-Series

Maybe i find some time tomorrow and can test the x710 with some SFP+. But i need to check which SFPs i have for testing in the lab. I give you an update tomorrow. bg Gregor

SFP+ CR Cables

I have the same NIC in our blades and they are working out of the box. root@pmxn1:~# lspci | grep Eth 01:00.0 Ethernet controller: Intel Corporation 82575EB Gigabit Backplane Connection (rev 02) 01:00.1 Ethernet controller: Intel Corporation 82575EB Gigabit Backplane Connection (rev 02) 02:00.0...

http://ark.intel.com/de/products/81825/ I would use the Intel Server with 2680v4 and today not less than 256GB RAM

Wieso brauche ich pro VM ein Volume das ist einfach nicht richtig! Ich erstelle einfach einen Flexclone vom Volume und mounte den Clone mit NFS und kann die VM einfach hochfahren. Ob auf dem Volume noch 20 andere VMs sind ist egal. Wieso sollten Snapshots keine Backups ersetzen? Man kann ja...

Warum nicht eine NetApp? Teuer ist immer relativ. Man kauft eine NetApp auch nicht für nur NFS. Eine NetApp kaufe ich wenn ich Funktionen haben will. Da ist vor allem die Sicherung mit Snapshots einfach wesentlich besser als ein herumkopieren bei dem ich Stunden brauchen würde um so etwas zu...

Please show me some tests where you get 100GBit/s. I don't know any storage system which could handle this traffic now. You need a lot of CPU to handle this. As i know you can do about 1M IOPs for now. Do you know any storage system with 100G interfaces?

100GBis is nice but i don't know any storage which can write this speed to disks or ssds. So this is useless in my eyes.

What performance you need and how many space do you need? We did tests for DR. They are working but you need some time get all up and running. Lets say 1-2 hours. Depending on your infrastructure. If you script this maybe 30min. The cheap FAS are 25xx. They are ok but limited in connections...

If you like to use the features you need to buy a FAS. The E-series are only a very fast block storage. We use a NetApp MetroCluster so write the date synchronous on both datacenters. We backup the data from the MetroCluster to the nearstore. We can switch between the datacenters without any...

NetApp Snapshots has their own products for creating consistent Snapshots. These products are available for Exchange, SMSQL, SharePoint and Oracle. There is also a software for VMWare. As i know they have support for RHEV also. This is KVM based so it should be possible to do this with Proxmox...

I would use NetApp as Storage and do Backups with Snapshots and use SnapMirror or SVM DR to backup site. This would be the fastest solution. SnapMirror is using network compression.

If you have too many broadcasts you have a too big network. Today i would use 2x10G for each server this should be enough for most scenarios. For testing or systems which does not need high performance you could use 4x1G in LACP. NFS/TCP tuning is very important search in the forum i posted...

Haproxy also supports https backends so why not doing it the right way?

Why? This a big security risk!!!