Zorlack's questions

I have a Kubernetes cluster running in EKS (on AWS.)

In the cluster I have Elasticsearch, Kibana and various other web services.

I would like to set up a single ALB loadbalancer such that:

• Requests to /_kibana/* is routed to service: search-kb-http, port: 5601
• All other traffic /* is routed to service: web-service-locator, port: 5000

So I create this:

#Note: search-kb-http is provided by Kibana operator
---
apiVersion: v1
kind: Service
metadata:
  name: web-service-locator
  namespace: default
spec:
  ports:
  - name: http
    port: 5000
    protocol: TCP
  selector:
    company-app: web
  type: NodePort
---
apiVersion: networking.k8s.io/v1beta1
kind: Ingress
metadata:
  annotations:
    alb.ingress.kubernetes.io/load-balancer-name: [SNIP]
    external-dns.alpha.kubernetes.io/hostname: [SNIP]
    alb.ingress.kubernetes.io/ssl-redirect: '443'
    alb.ingress.kubernetes.io/certificate-arn: [SNIP]  
    alb.ingress.kubernetes.io/scheme: internet-facing
    alb.ingress.kubernetes.io/target-type: ip
    kubernetes.io/ingress.class: alb
  name: public-ingress
  namespace: default
spec:
  rules:
  - http:
      paths:
      - path: /_kibana/*
        backend:
          serviceName: search-kb-http
          servicePort: 5601
      - path: /*
        backend:
          serviceName: web-service-locator
          servicePort: 5000

Here's the problem

I prefer to do the SSL termination at the ALB for the main web-service. But the Kibana backend only offers HTTPS. So I need the two different backends to use different protocols in transit.

In doing my research it seems there is an annotation I can apply to the ingress:

alb.ingress.kubernetes.io/backend-protocol: HTTPS

But this changes seems to affect the entire ingress.

Is there a way to specify different transit protocols for different backends?

Many thanks!

[EDIT] I have found a workaround to disable TLS on Kibana in ECK. But the question still stands as to whether this is possible. Thanks!

I've been using httping to monitor the performance of my load balancer.

The httping utility accepts a -S argument which is briefly described in the man page:

-S Split measured latency in time to connect and time to exchange a request with the HTTP server.

The output looks like this:

connected to [hostname-snip]:80 (284 bytes), seq=259 time= 0.01+ 21.96+ 23.95+ 0.07+ 0.02= 45.99 ms

From this its clear that httping is breaking out the timing into Connect/Transmit/Wait (etc...) and also showing the cumulative output, but I can't account for what each of the values mean.

Does anyone know what each of these values mean?

I recently built a small cluster for running Solr. The cluster consists of 12 Supermicro Blades running E3-1270V2 with 32GB of ram.

11 of these servers are running fine. One of them crashes on me constantly. When the server crashes it typically produces some output on the terminal. The first time it was:

double fault: 0000 [#1]

Hmm... thats pretty cryptic. Since then I've recreated the problem and gottem some more interesting messages.

Here's another equally cryptic message...

Another interesting wrinkle is that I can fire up sysbench and max out the CPU without it crashing, but it's not until I start Java that it crashes reliably.

I've tried turning off the following CPU features:

• Turbo Mode
• C States
• T States
• XHCI

Is this just a bad CPU?

Many thanks!

I have a Supermicro 2uTwin2 blade server which we're hoping to run XenServer on.

Because this motherboard only provides Intel's fakeraid, I was hoping to leverage its boot from iSCSI functionality to get some redundancy.

I can't figure out how this is meant to work.

My best guess would be that the NIC acts like an HBA, allowing me to boot off a CD and then install the OS onto the NIC-based disk. However this doesn't seem to be the case.

I can define an iSCSI target in the BIOS, and when the server boots I see it connect momentarily. Then it seems to disconnect and move-on.

So, my questions:

• Can someone explain how the Boot From iSCSI process is meant to work?
• How can I install an OS on to an iSCSI volume?

I have a Synology RS812RP+ which I'm using for a little bit of departmental testing. I've stuffed a bunch of SSD disks in it, and I'm curious about monitoring the disk IO.

iostat doesn't come pre-installed on the synology and it doesn't have the tools to build it. Does anyone have any advice about getting iostat running in this environment?

We have a XenServer pool consisting of two Dell R410's running XenServer 5.6. The servers are connected to a few iSCSI NAS devices.

Lately, as I've been looking at the /var/log/messages I've been seeing a lot of this:

Apr  6 12:01:27 xenserver-alpha xapi: [error|xenserver-alpha|345 xal_listen|Event thread updating guest metrics (domid: 87) D:7cfbb800692e|sql] Duplicate key in set or map: table VM; field platform; ref OpaqueRef:b6808180-fabd-07c7-d19c-df3e345e8854; key viridian

I'm not sure what this error message means. I've seen it on both XenServers (lets call them xenserver-alpha and xenserver-beta).

Googling as brought me a few posts on the XenServer forums, but none of them seem to answer my question, which is:

What does this error mean?

Should I be concerned about it?

How do I fix it?

We have an application running on Classic ASP which I'd like to get to use connection pooling.

The application currently uses this as its connection string:

"Provider=MSDASQL; Driver={SQL Server}; Server=db.example.com; Database=DBName; UID=Username; PWD=Password; ConnectionTimeout=15; CommandTimeout=120;" 

When I bring up the ODBC connection pooling perfmon I don't see any connections in the pool.

I've read this article which suggests setting a value in HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\W3SVC\ASP\Parameters but there's no \ASP key so I'm not sure if it applies to IIS 6.0.

Does anyone have experience with this issue? Is there a simple way to toggle pooling on?

On a daily basis we generate about 3.4 Million small jpeg files. We also delete about 3.4 Million 90 day old images. To date, we've dealt with this content by storing the images in a hierarchical manner. The heriarchy is something like this:

/Year/Month/Day/Source/

This heirarchy allows us to effectively delete days worth of content across all sources.

The files are stored on a Windows 2003 server connected to a 14 disk SATA RAID6.

We've started having significant performance issues when writing-to and reading-from the disks.

This may be due to the performance of the hardware, but I suspect that disk fragmentation may be a culprit at well.

Some people have recommended storing the data in a database, but I've been hesitant to do this. An other thought was to use some sort of container file, like a VHD or something.

Does anyone have any advice for mitigating this kind of fragmentation?

Additional Info:

The average file size is 8-14KB

Format information from fsutil:

NTFS Volume Serial Number :       0x2ae2ea00e2e9d05d
Version :                         3.1
Number Sectors :                  0x00000001e847ffff
Total Clusters :                  0x000000003d08ffff
Free Clusters  :                  0x000000001c1a4df0
Total Reserved :                  0x0000000000000000
Bytes Per Sector  :               512
Bytes Per Cluster :               4096
Bytes Per FileRecord Segment    : 1024
Clusters Per FileRecord Segment : 0
Mft Valid Data Length :           0x000000208f020000
Mft Start Lcn  :                  0x00000000000c0000
Mft2 Start Lcn :                  0x000000001e847fff
Mft Zone Start :                  0x0000000002163b20
Mft Zone End   :                  0x0000000007ad2000

Today we diagnosed the cause of a Periodic slowness issue: see here.

Dell OpenManage Server Administrator seems to have been causing hourly slowness. It would occasionally peg one of the CPUs for upwards of two minutes. Disabling it drastically improved the performance of the SQL Server.

The server hardware:

• Dell R710
• Dual Quad Core 2.9GHz Processors
• 96GB Memory
• 2 Disk RAID 1 SAS System Disk (Internal)
• 4 Disk RAID 10 SAS Log Disk (Internal)
• 14 Disk RAID 10 SAS Data Disk (External DAS MD1000)
• Windows 2008 Enterprise R2 x64

We installed the OS using Dell OpenManage Server Assistant, so I assume that it was correctly configured.

For now we have disabled OMSA to alleviate the performance issues it was causing, but I'd like to be able to re-enable it.

Has anyone had a similar experience that can shed a little light on the nature of this problem?

We're trying to diagnose the cause of slowness on our Database server.

We're running the latest rev SQL Server 2005 on Windows 2008x64.

The behavior that we're seeing is this: We see the SYSTEM process spike one of the CPUs for about 2 minutes, during this time SQL server slows down by a factor of 10. The slowness lasts until SYSTEM is done, then in an hour everything starts again.

During these slowdowns disk writes don't spike, paging doesn't spike, the only noticeable precursor we see is that SYSTEM maxes out one of the sixteen (HT)CPUs.

Note that this doesn't happen at the top of the hour, it just happens once an hour, and it shifts a bit depending on the length of the incident.

At the moment this is causing intermittent slowdowns, but when the server is really busy it can cause Worker Thread starvation.

The server is a Dual Quad Dell R710 with 96GB of RAM and RAID10 data/log disks.

Has anyone experienced this kind of problem? Does anyone know where we should look?

Edit: SQL Server Version is 9.0.4035